Jump to content

Experiences of a PCV with Autotune on a 2004 VFR800


Guest GPM

Recommended Posts

Warning - Long post full of technical stuff combined with opinion ahead. You were warned.

***

Background

I was actually quite a happy-chappy with my VFR - the mods I had performed had already smoothed out the VTEC transition (though it wasn't bad to begin with), and response down low wasn't too bad either. My main complaint, if I was to have one, was the abruptness of the throttle control in 1st gear which made low speed work a chore (and I would normally cheat by feathering the clutch). I decided to grab a PCV (and eventually autotune) partially because it would hopefully make the bike even better at those infamous points, but also inspired by CodeRighter's efforts I decided 'why the hell not'. I guess in part this post is part of that why-the-hell-not process wink.gif .

I decided to buy my PCV and Autotune from the USA rather than locally - yeah, I know, frowned on by the power that be (Dynojet) but the reason was very simple - price. I don't have unlimited funds to throw about, and frankly the cost of purchasing those little boxes locally is crazy. On a good day I was looking at $AUD550 to 600 for the PCV, and another $450 to 500 for the Autotune. That means if I was getting it as cheap as I could I would be looking in the region of $AUD1000. In recent currency terms (with the $AU buying about 90 US cents) I was look at about $USD 900 if I bought locally. If I bought from the USA I was looking at about $USD 500 plus another $40 in postage (or $AUD 600, a $AUD 400 saving!). So in order to save myself some significant cash I obviously took the order from the US route. Yeah, yeah, no warranty or official support - but both parts would have to fail (and I order replacements) for me to only just be out of pocket compared to if I bought locally. I am all for supporting local business, and normally do so when possible, but I don't support price gouging by importers with an exclusive deal. I recommend to anyone else who would benefit from significant savings by buying internationally to do so smile.gif .

As I was going to order from the US and I was in a penny pinching mood I ordered the PCV first and the Autotune about a month later (basically to pick up the items when there was a good deal on each of them). This meant I ran the PCV for over 4 weeks just by itself, and that alone taught me a few interesting things which I shall share shortly. As such, I will also break this part up into a PCV section and an Autotune section (hopefully that will also be more helpful to those not interested in the Autotune).

Going into this modification bonanza I had made a few relevant alterations to my bike (most of which are pretty much universal).

  • Disabled the PAIR system
  • Removed snorkel
  • Disabled flapper
  • K&N air filter

Otherwise, from an engine and mapping perspective, she was stock (so it still had the original exhaust).

PCV - Install and setup

Installation of the PCV (and PCIII) have been covered elsewhere, and as those guides point out, it's very simple.

All you really need to do is remove your rear seat to find some place to put the little red and black box (being non-ABS I placed mine behind the rear brake fluid reservoir, but it is small enough that you can place it in a number of good spots). Then lift your tank, fish out the designated cable connection (should be just behind the frame on the right hand) - disconnect this connector and plug it into the PCV. Now install the supplied O2 eliminators on the left hand side (you will likely need to remove the fairing for this). I tried to ensure that the now exposed O2 sensor plugs were still shielded inside the rubber electrical connector protector, but you could tape them up for extra peace of mind. After that you attach the ground wire to the negative terminal on the battery and you are set. Luckily the device comes with a pictorial instruction set showing you exactly how to do this, so you really cannot go wrong.

IMG_2139.jpg

Where my PCV lives

Once you have done that you are good to go. There are other wires you will need to hook up for more advanced features (in particular tapping into the speed sensor for map-by-gearing and the temp sensor for... well, engine temp. but I will get to those later as their value is limited without the autotune).

Hookup your PCV to your (Windows) laptop, load the software and it should automatically pull down the map pre-loaded onto the device. Now things alter a little for the 2002-2005 bikes (as the map comes preloaded with the 2006+ map). You will need to upload a new (suitable map) before you take it for a test drive, mostly due to the VTEC transition point changes (and probably a few other tweaks Honda implemented on the 2006+ ECU). With the latest version of the PCV software this is a piece of cake - simply download the suitable PCIII map and use the inbuilt converter - easier than doing by hand (as I had to). Once that is uploaded you are good to go! You may also want to check your throttle position settings and give it a run on the centre stand with your laptop plugged in, just to check that everything is working correctly.

PCV - Usage report (or, WTF is this thing doing!?!)

I personally decided to play it conservative and loaded up the default Dynojet 2002 map. I planned on eventually giving Cozye's map a go until I obtained an Autotune. To my surprise, and horror, the default map ran like absolute crap. It's hard to find words which described the horror of it's operation, but it was shocking. Surging all over the place, bucking and general bad, bad running. Realising that when diagnosing fuel mixture problems it's 'better' to go too rich then too lean, I decided to richen up the map (theory being that if it got worse, or failed to improve, I would lean it up instead - I had noticed the default map was considerably richer then Cozye's). This helped the problem, it was better behaved (but still clearly unhappy), but as you can imagine, fuel consumption went to the dogs. It had me stumped, the bike was fine before I installed the PCV, and the only other change at the time was the O2 elims (which are needed as per the Dynojet instructions). Clearly there was too much air in there, causing it to run lean. The only other clue I had was it was much worse below 6000RPM then above.

Resolved to solve the issue, I decided to start restoring parts of the bike to stock to see if some other change had caused the problems. Seeing as there are not many relevant changes made to my bike I decided to do it in two steps - first the flapper and snorkel, if that didn't work, the PAIR system and air filter (I planned on returning the air filter to the OEM model eventually anyway, but I'll cover why at a later point).

First up, flapper and snorkel. Up goes the tank and I man-handle those two back to their original function. I take the bike for a test ride and suddenly - wow, fixed! Still a little grumpy at me, but night and day type change from before - I safely put the remaining grumpiness down to the current map just being a default one rather than a map tweaked or custom made for the setup. My theory is that the flapper was at fault (because frankly the snorkels impact would be small compared). With the flapper disabled the bike would pull in too much air below 5500RPM (which from memory is it's cut-off, or rather, fully open, point), which when in closed loop mode would be compensated for by the ECU. Eliminate those O2 sensors and suddenly it doesn't work so good any more. Of course, this doesn't make sense in light of the number of people out there who have no issues with their PCs or O2 elims with the flapper disabled, but it did the trick for me.

So the lesson there is - if your bike is running like crap after you have just installed a power commander, check you air intakes.

After restoring the flapper I loaded up Cozye's map, which run rather well really. Over the next few weeks a tweaked it by feel, and the map progressively got better and better (well, felt like it anyway). I decided at that point that PCVs are great, and keenly awaited my next new 'toy', the Autotune.

Autotune - Install and setup

As I said above, I obtained my Autotune a little over 4 weeks after I had installed the PCV. By that point I was happy enough with my tweak-by-feel version of Cozye's map - enough that it would make a good base map to start with for my Autotune adventure.

Installation is pretty easy again, but there is a bit more to it then for the original PCV. As long as you have the tools it should not take you more then 90 minutes (including drinks breaks) if you take it slow and check you work. May I suggest you take it slow and check your work (I'll tell you why shortly). The instructions are pretty good so you should follow them, but they are not as good as the instructions for the PCV unit (I guess that's because they are generic all-bike instructions).

A few tips:

Do not forget the CAN terminator - you put it in the spare plug on the Autotune, not the PCV.

Make sure you install the cables properly - on the Autotune module there is a membrane protecting the internal connectors which you have to pierce - I used a small jewellers screwdriver first to do this, which made inserting the cables a piece of cake.

Make sure you get the cables in the correct order.

You can install the sensor in either of the O2 sensor positions (one is for cyl 1 & 3, the other cyl 2 & 4), but you will have a much easier time of it installing it in the leftwards most one (see pic).

I thought the O2 sensor cable connector (used to plug the sensor into the extension that runs to the module) looked like it was asking for dirt and crud to get into it, so I cocooned it is electrical tape. Probably not necessary, but I would rather not spend a few hours down the line cleaning it out with compressed air and contact cleaner (see pic).

I suggest hooking up the speed sensor and temp sensor wires now. Save yourself the trouble of doing it later (even if you don't intend to use either, though you really should). A decent length of wire will suffice, and use either posi-taps or slip the wire into a connector trick to connect them. See the below pictures to see which wires you are looking for on the bike. Remember these go to the PCV itself, not the Autotune.

IMG_2140.jpg

My Autotunes temporary accommodation (due to cableing for now)

IMG_2146.jpg

Where the O2 sensor goes (or rather replaces the existing one). Yes, my bike is dirty and in need of a good clean

IMG_2150.jpg

Autotune cable (in the black sheath) and wires for speed and temp sensors (white and yellow)

IMG_2152.jpg

Temp sensor is the yellow-with-blue-stripe wire

IMG_2154.jpg

Speed sensor is the pretty little pink wire

IMG_2153.jpg

O2 plug wrapped up for its own protection

IMG_2155.jpg

Everything nicely tucked away

Once I had everything install, I popped the bike up on the centre stand, turned it on and let the bike warm up with the PCV plugged into my laptop (so I could see all the readings). Much to my horror, the AFR continued to be reported as 9.99 - according to the instructions this is a bad sign. I went over and double checked everything but it still reported 9.99 - but no other errors and its little LED was blinking correctly. Turns out that before I wrapped up the 12v connection for the Autotune (which powers the O2 sensors heating element) in electrical tape, I forgot to actually connect the lead to a 12v source! (whoops). Once I did this everything went fine. Lesson here is make sure everything is correctly hooked up - and no heater, no working, even though the exhaust would have gotten it up to operating temp.

Final setup is easy too - Dynojet supplies a default AFR map which you can use (and after taking a quick look it should be fine to use the 2006+ AFR map on a 2002-2005). I cheated and recycled CodeRighters existing good work, which I then altered to suit my needs (basically leaning out where I tended to cruise). Activate the Autotune functionality in your PCV software and you are good to go (I'll cover gearing and temp shortly).

Autotune - Usage report (without by-gear mapping)

For a couple of reasons, the main one being the KISS principle (ie. keeping it simple) I decided to start off without using all that fancy PCV stuff like map-by-gear and accelpump. I figured I would get a handle on the basics first, and then move on to the advanced features. Let me summarise the rest of this section for you - don't bother with that crap, just skip straight to map-by-gear and set the rest up right to begin with - yeah, sure, there is more points of failure where a mistake could cause problems, but it will save you a lot of screwing around otherwise. There are now a few who have blazed this trail before you (and CodeRighters write ups are an excellent resource, so use them) so you can just copy what others have done.

Now for the detailed information. The Autotune worked, it actually worked rather well - considering that it was not made for the 2002-2005, nor supported, I was a little surprised at how easy it all was (sure, there isn't that much different between 2002-2005 and 2006+, but there are key differences, enough to cause problems). I started riding, it started trimming, I started accepting these trims and letting the Autotune work towards a nice balanced map of its own. The trims were pretty mad to begin with, but I sort of expected that - I figured it would work itself out over time (basically averaging its way towards an approximate optimum).

To help demonstrate what’s going on each time you go out for a ride (and then accept the new trim tables) here are some nice pretty graphs coming some sets of readings. For reference there is an average of CodeRighter's map (which he made using his autotune over some considerable time), there is the base map which I started with, and there is the fuel map from a number of 'reading' points were basically I have accepted the trims after a decent amount of riding (normally in the region of 100km). I have only included 2% to 20% throttle maps as they show the greatest amount of change (it starts to get less and less after that as I tend not to ride around at 60% throttle much).

2%25.jpg

2% throttle

5%25.jpg

5% throttle

10%25.jpg

10% throttle

15%25.jpg

15% throttle

20%25.jpg

20% throttle

Graphs are great to see what sort of changes are going on, and clearly show the Autotune adapting as you ride (though you can tell that through feel as well). Unfortunately what these graphs don't is how it feels, and let me tell you, it felt like crap. Absolute crap. Similar to when I first installed the PCV, but in a different way. At points it was snatchy, at others, it was surgy, others boggy and occasionally bucky. Just bad. The real bad part of it was that it would switch between good and bad over the course of a ride (so obviously the system was trying to fix everything up).

I'll try to describe exactly what I felt was going on, and that should also make my eventual solution clear. I would start off and everything would be going fine, Get up into 2nd (which I tend to use as my suburban speed gear, so 40-60km/h in built up areas) and move off from home, all is good. Shift up to 3rd which I use for normal use 60-90km/h (and if I want some VTEC action 100km/h+), and things would feel bad, a little boggy, unresponsive, but it would fix itself shortly. Continue onwards basically switching between 1st, 2nd and 3rd (so suburban riding ranging from streets to main roads to avenues). Over time, it would get worse and worse and worse, and always at its worst just after a gear change - so I would be sitting at a particular RPM and throttle position for 2nd gear, then a short time later, same RPM and throttle position in 3rd gear, and it would run like absolute crap (sometimes even to the point where it felt violently unsafe). I would then later switch back to the same RPM and throttle position in 2nd gear (as I move between roads and streets) and it would be running again like absolute crap. Normally it would resolve itself, but it tended take a couple of RPM 'sweeps' (basically open the throttle a little to increase the RPM, close it a little to drop it below the range in question, then re-open, and re-close, do this 3 or 4 times and the feel improved, but wasn't really fixed). The solution was clear, map-by-gear - clearly the setting that the Autotune was establishing for a particular RPM and throttle position combination would work in one gear, but not in another. You can sort of see the problem in the graphs where between one reading and another there has been a crazy change - which I took to mean that in one trim-accept the last time I was at that throttle and RPM point I was in one gear, and at the other trim-accept I was in a different gear.

Lesson here follows, just go straight to map-by-gear. Really, it is that much better.

Autotune - Usage report (with by-gear mapping)

Map-by-gear is pretty easy to set up. There is no gear indicator on our bikes so we need to basically tap into the speed sensor and let the PCV do some quick calculations to turn speed and RPM into what gear we are in. It's actually pretty simple.

First you need to tap into the speed sensor. Basically hook the correct wire port on the PCV up to the pink wire coming out of the speed sensor (see the picture I took for where I tapped in). That's it from a physical point of view. I'll get to the software side shortly.

While you are at it, set up the temp sensor (so that you can tell the PCV to hold off using the Autotune until your engine is up to spec.). You will need to tap into a yellow-with-blue-stripe wire not too far away from the infamous blue connector. This also goes back to you PCV. I just routed the wires for these two back under the tank (following most of the other electrical bundles flowing through there).

Once that is done you need to set everything up in the software. My PCV came pre-configured with the speed scalar and gearing already set, but nothing in temperature. The problem is I live in the Rest of the World, and we here in the Rest of the World use the metric system (because it's frankly easier - decimal number system combined with a decimal measuring system) and the device was set up for those strange measurements you use over there in the USA. So I decided to reconfigure it all to make my life easier - the PCV doesn't care on what measuring system you use (if you want to measure your speed in parsecs, go right ahead), so long as all the different numbers used in the maths match up.

For those interested in the figures I came up with... (keep in mind these are for METRIC measurements, some may also be applicable to imperial applications, but you are better off doing it for yourselves)

Scalar.jpg

Interestingly if you take the speed scalar I calculated and multiply it by 1.6 you get a figure very close to what Dynojet supplied (a little over 9000). This is what you would expect and shows that both Dynojet's scalar is probably on the money and the one I came up with is good too.

Gearing is the next important thing. If you don't want to use the Dynojet supplied version you can do it yourself, which is what I did. Just pop the thing on the centre stand, turn it on, put it into gear and off you go (without, obviously, going anywhere). This is much easier if you have a second person to capture the gear information on the laptop, but I was able to do it solo.

For those interested, here is what I came up with.

Gears.jpg

If you want to check that your settings are good it is a pretty simple piece of mathematics. Find a speed you like to travel at and make a mental note of the gear and RPMs. Next time you have a calculator handy take the RPM, divide by speed and then see if that number you come up with fits between the high and low gear points it the PCV. If it does, good stuff, everything is as it should be. If it is pretty close to the edge you may want to recalibrate the gear ratios for the PCV, but that's up to you.

As an example, say in 3rd gear I tend to be a little over 4500 RPM while travelling at 60km/h (which is about what my bike reports, but this is an example). If I go 4500/60 I get 75. Looking at my gear calibration in the PCV 3rd gear is about 82 to 65 - and 75 is well within that, so it's about right.

That's a pretty quick and dirty way of doing it, but if something is not right and you are using map-by-gear, it is worth having a quick check of such things.

Temp sensor isn't hard to configure either. I did a rough and ready setup (in Celsius, again metric folks), it’s probably a degree or 2 out at various points but extreme accuracy is not really needed as I only use it so the PCV knows when to start listening to the Autotune (which I do at 70 degrees C).

Temp1.jpgTemp2.jpg

Now onto the main show, the usage report.

Thankfully, map-by-gear saved the day. It really did fix the problems I was having before and the Autotune started to shine (which is good, the damn thing wasn't cheap!). Once you have this thing down pat it's worth spending some time tweaking - after reading some posts by CodeRighter I started playing with the accelpump function and can report the same findings as he did, setting it so that the PCV would ignore the autotune (and letting the ECUs accelpump functions work their magic) made it feel better under accelleration. I personally set mine to have a delta of 80% and for it to last for 100 revs with no increase in fuel (so it basically takes whatever the ECU says should go happen + the existing fuel map in the PCV). Also take the time to tweak your AFR table, and whenever you accept the trims, take a look through the table for any obscene changes from one cell to its immediate brother (they do slip in there, more on that shortly).

Performance is great, the whole bike feels better now, especially as I have been able to progressively refine the autotune generated fuel map (though riding!). Fuel consumption has dropped considerably as well, and I get some strange enjoyment out of taking a look at progressive graphs of changes in the fuel map (maybe that's my inner science/maths nerd feeling good about all that data!). I'll post up some more graphs from the map-by-gear setup shortly, but I have to work out which ones are worth looking at (as obviously there are many, many more graphs then pre-map-by-gear - about 6 times as many!).

The main issue I have encountered is when the Autotune decides to be silly with neighbouring cells. As in, say at 5% throttle, 5000 RPM you will have +15 in the fuel map, and then in 5250 RPM you will have -15 in the map. The Autotune, I think, allows this when you are decelerating (sudden rush of air from the decel vacuum confuses the O2 sensor and makes the system believe it is running really lean, so it richens up that cell). This problem should be resolved next time you pass through that cell on the way up, but sometimes that may not happen for a while (typically in a RPM/throttle/gear combination you do not experience frequently). This only affects the system for a very short period of time, basically that period before you fully close the throttle (as the autotune is ignored at 0% throttle), or when your RPM levels out if you are only partially closing the throttle. It does become a nasty problem in one set of circumstances, if you are cruising along with a steady RPM and throttle position, that just happens to be close to the transition point between 2 cells of significant difference, you can enter this sort of self supporting cycle of cell-jumping - basically, you enter into the 'bad' cell, the engine is suddenly starved of either fuel or air, as such power drops and with it the revs, they drop down into a 'good' cell, starvation is resolved, engine happy, revs jump back up into the bad cell, and so on. Very annoying, and as the Autotune is a post-mortem device, it doesn't seem to be able to pick up the problem very well. The only solution I have found while out on the road is to shut the throttle, let the RPM drop below the good cell, then power back on (gently, you are aiming for the problematic throttle opening AND RPMs) and let the bike rev up to above the 'bad' cell. Do this 2 or 3 times, problem solved. Another way (though it takes a little longer) is just to increase the revs in the 'bad' cell so that it is above the transition point, the Autotune will then resolve it on its own (probably as it is getting a stable reading from the O2 sensor). The problem would probably go away if Dynojet implemented a 'reverse' accelpump (a decelpump if you will) that would tell the PCV to ignore the autotune over a particular throttle-closing delta.

And as I promised, some graphs (with analysis) for Autotune with map-by-gear. I have tried to select 'interesting' graphs (because that means alot less graphs to save, upload then embed!), but if anyone wants a particular comparison done, just let me know.

nb. I have hidden it all in spoiler tags in order to save a bit of scrolling.

1st gear

Less then 10% throttle 1st gear is very erratic - I put that down to the fact that at those sorts of throttle positions I am powering up rather then level cruising, thus the fuel map will change frequently due to fuel, temp, and just how I feel like riding that day.

10.jpg

Here CodeRighter's map is very spikey, mine is smoother - though there seems to be an interesting attraction to enriching 4750 RPM.

15.jpg

Things remain relatively smooth through the line, but the Autotune clearly wants to lean out between 3000 and 6000 RPM.

20.jpg

At 20% it seems Autotune wanted to stick to around the default map (which is basically Cozye's) - this remains the same at higher throttle openings - where Autotune has made a change it hasn't been dramaticly different from Cozye's map.

2nd gear

I ride alot more in second gear then first, and it shows.

5.jpg

Common theme here seems to be enlean-enrich-enlean-enrich. That there are not dramatic changes between readings is a good thing, the changes are more or less stable. The pattern seems to match up with where I am winding up the revs to get to speed (or when coming out of a slow corner), so I am not staying in 5% for long, then it leans out where I would be cruising, then starts getting rich again when I am wanting to stretch the bikes legs.

10.jpg

A pretty uniform (if a bit spikey) progression. A strange contrast to the 5% throttle graph.

15.jpg

Cozye's map seems to be way too rich here, and CodeRighter's map agrees! It does follow on from the 10% pretty nicely though.

20.jpg

At 20% things start to richen up again. Interestingly Reading 6 there was straight after a very spirited ride in the hills and twisties, yet it is wanting to be leaner then the others.

40.jpg

At 40% things start to look a bit more uniform, but I don't spend much time in that area when I am going around town, so the chances to get a reading are reduced.

3rd gear

2.jpg

2%s tend to be very erratic as I tend to be either opening or closing the throttle at that point (rather then allowing time for a refined trim while holding the throttle steady). Even so, this graph is interesting as you can spot clear trends at particular RPM points.

5.jpg

An interesting map due to how much change there is from one reading to another! Also a few interesting rouge spikes (which I think is related to the decel issue). You can also see at 5750 RPM from Reading 4 to Reading 5 that if you don't get back to that point in the map the 'bad' cell data can stick around for a while (we are talking 150-200 km here between readings!). Reading 6 eventually smoothed it all out though.

10.jpg

10%, with a few exceptions, is much smoother then 5%. But it also once again shows just how rich Cozye's map is (and CodeRighter's map seems to agree).

15.jpg

My map wants to remain leaner longer then CodeRighter's, but both again indicate Cozye's is too rich. I would put the differences between CodeRighter's and my maps as differences in mods (as I have my bike currently as stock as possible in the air/fuel department).

20.jpg

Don't know what happened here - flat but jumpy. Seeing how closely I follow CodeRighter's line I guess it is meant to be. Honda's map must be interesting though...

4th gear

5.jpg

This graph is good because you can clearly see the decel issue at 5000 RPM. Now imagine sitting at about 5000 RPM, bike is starved for air, looses revs, drops to 4900, suddenly has a higher air to fuel ratio, jumps back up (due to now normal power), and cycle. Clearly the 5000 RPM reading is an error.

10.jpg

And contrast here where it's gone. Likely what I did is decel through the 5% position (without closing the throttle), but the 10% zone is fine.

5th gear

5.jpg

Spikes everywhere - I think due to small decel while cruising (basically small speed adjustments on highways due to traffic, gentle slopes, etc).

10.jpg

My and CodeRighter's maps seem to agree (and again disagree with Cozye's).

20.jpg

Haven't used much 20% in fifth, but still an interesting display of how the Autotune makes changes over time (and doesn't always stick to them in the long run).

And for those interested, here is my map (as at Reading 6). Note that it is not a mature autotune reading, it has had in the region of 1500km of travel at this point. Use at your own risk, etc.

Autotune - Looking to the future

As it is all running well, I am starting to turn my mind to what to do next with my Autotune. I am thinking of running some experiments with it - mostly to see what sort of effect many of the common small mods have. One thing I would like to try is to disable the flapper valve again and see what the Autotune thinks (ie. will it richen up the mixture due to the extra air?). Also I wonder what changes to air flow my K&N filter has (which is why I put the OEM back in). Never know, I may even be able to spot some evidence that those budget mods to have an impact on performance (though the scale, and whether it is positive or not would remain unknown). I'll have to think about all of that.

I am more than happy to help out anyone else thinking of installing (or having trouble with) a PCV or Autotune. If you have questions, or want pictures, feel free to ask!

For those interested, screenshots of my throttle position and Autotune settings.

Throttle.jpg

Autotune.jpg

(and I did warn you this would be a long piece!)

And now to finish off, various pics which may be useful...

IMG_2157.jpg

Where the speed and temp wires go (this is set out in Dynojets instructions)

IMG_2159.jpg

Labled image of the device in its housing - note that the USB cable is there on a permanent basis - I have it set up so that I can remove the USB type A plug from under the seat (and put it back again) without having to remove the seat at all - a real time saver!

AllWorking.jpg

A screen shot of the PCV software with the bike at idle (and having started to warm up)

IMG_2160.jpg

A shot of it working, and you can check the software figures against the dash!

Edited by GPM
Link to comment
Share on other sites

  • Replies 138
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

This is actually the first time I've seen a this detailed report or even paid this close attention to how in depth the autotune is. I'm impressed. Look forward to hearing more about your experience as your weeks go by, I may have to see about doing this myself, just to have a hands on and see the comparison is after riding with PC3 for years. Great report BTW, Thanks.

Link to comment
Share on other sites

  • Member Contributer

So, what target AFR values did you use? 13.2 across the whole range?

Did you just take Cozye's PCIII map, convert it to a PCV map, enable Autotune and leave the stock AFR values?

Whoops, I found what you did:

Final setup is easy too - Dynojet supplies a default AFR map which you can use (and after taking a quick look it should be fine to use the 2006+ AFR map on a 2002-2005). I cheated and recycled CodeRighters existing good work, which I then altered to suit my needs (basically leaning out where I tended to cruise). Activate the Autotune functionality in your PCV software and you are good to go (I'll cover gearing and temp shortly).

So that was good enough, eh? Hmmm. I still think a dyno tune which finds the best AFR for performance on a particular bike would be best.

Can you please post your map file for me to download and have a look at?

Link to comment
Share on other sites

Man, that's a lot of fiddling. :rolleyes:

Hehe, my better half once commented that my bike's maintenance has a lot in common with a jet fighter. When I asked why she told me it was because I seemed to spend 100 hours working on the thing for every hour I rode it.

I do a lot of fiddling wink.gif

So, what target AFR values did you use? 13.2 across the whole range?

Did you just take Cozye's PCIII map, convert it to a PCV map, enable Autotune and leave the stock AFR values?

Whoops, I found what you did:

So that was good enough, eh? Hmmm. I still think a dyno tune which finds the best AFR for performance on a particular bike would be best.

Can you please post your map file for me to download and have a look at?

I used Cozye's map for the PCV to begin with, and tweaked it as appropriate. Not huge changes, but just fixing up obvious problems (normally boggy points or hesitation). Then I used that as my base map for when I got my Autotune.

After that, as you read, I used CodeRighter's hard work for the AFR side of things. I believe he just tweaked the default anyway, but he has been running it for a while and his bike hasn't exploded, so I guess that's as good an approval as anything. I played with it a little myself (mostly leaning out where I tend to cruise for fuel efficiency).

A dyno tune set-up AFR map would probably be good - but hey, that's more money I don't really want to spend. AFR tweaking is a lot easier then fuel map tweaking - with the fuel map you are taking a stab hoping to get a particular outcome (AFR) - with the AFR map you are telling it the outcome you want. Seriously it's a matter of working out (from your own riding habits) what you tend to do at particular throttle positions at various RPMs (in which gear, if you map by gear), and changing accordingly. I am currently operating on a 13.2 for performance, 14.0 for max fuel efficiency range (and everything else in between). In most cells I still have 13.2, in the cells I tend to 'utilise' the most in day to day riding I have tweaked around. For example, if I am doing 60km/h in 3rd gear I am probably commuting or taking it pretty easy, so the 3rd gear AFR cells which correspond to this sort of speed I have made lean (13.6 to 14.0). 100km/h in 3rd gear however, I am looking at something a little more fun (and obviously dialing up some VTEC enjoyment) so the AFR would be around 13.2 to 13.4. The beauty of the Autotune with map-by-gear is this flexibility, I really can make the bike do all things in one package - from a super smooth, super quite late night suburban street ride (<4000 RPM), to a light on the pocket touring/commuting/cruising machine, to a dial it up to 11 twisty munching performer. And if I ever change my mind, its easy to quickly reset everything to what I want out of the bike! In my rather humble opinion I would rather be able to fiddle as I see fit to meet my needs rather than have a dyno tuner set my AFR tables for balls-to-the-wall performance from 0 to 12500 RPM. A good tuner who would be willing to put together a couple of different AFR maps for different performance requirements however... well, that would be interesting, just to see if there are any spots which are a little unusual to us amateurs.

I'll post up my map shortly, it's still a work in progress (hasn't had the maturing time that CodeRighters has, nor have I been able to ride recently - first loads of crappy wet weather, now I have a ruptured ear drum). You can analyse at will. I'll also post up more graphs to show how it all works.

Great read!

Didn't Codewriter use a vacuum solenoid to turn off the autotune on decel?

He did, and he reports it as being an excellent solution. Personally I think it shouldn't be necessary, Dynojet should be able to take decel into account software side. Failing that it is a minor annoyance really, and in the long run for me it will be a toss up between the effort required to fix using a vacuum switch and the frustration it causes.

Link to comment
Share on other sites

  • Member Contributer

Didn't Codewriter use a vacuum solenoid to turn off the autotune on decel?

He did, and he reports it as being an excellent solution. Personally I think it shouldn't be necessary, Dynojet should be able to take decel into account software side. Failing that it is a minor annoyance really, and in the long run for me it will be a toss up between the effort required to fix using a vacuum switch and the frustration it causes.

It also stops you from having two distinct maps, does it not? If you bond the switch to controlling the autotune, then it's not available for map selection.

I noticed my PCV eating a lot of memory just with a map-by-gear without Autotune enabled. What percentage is consumed when map-by-gear *and* autotune is enabled I wonder?

Link to comment
Share on other sites

It also stops you from having two distinct maps, does it not? If you bond the switch to controlling the autotune, then it's not available for map selection.

If you have a toggle switch AND autotune is enabled the toggle switch wont switch between maps. Instead it will switch between learning mode (ie. Autotune does it's thing) and the base map with the latest trims 'applied' (apparently). I don't know what the applied means, that is, if they are accepted or just added on, but that is what the literature says.

Sadly no switching between AFR maps on the fly. Would be cool if you could do that though (and give you even more customisation!).

I noticed my PCV eating a lot of memory just with a map-by-gear without Autotune enabled. What percentage is consumed when map-by-gear *and* autotune is enabled I wonder?

With map by gear and AFR by gear enabled and a 250RPM resolution, I have 54% usage. Yes, enabling map by gear really sends it skywards, I have effectively 3 maps for every gear (giving 18 maps in total!) - an AFR map, a trim map and a fuel map. Not sure how it would work with independant cylinder control enabled as well, but you would probably cut it close.

Link to comment
Share on other sites

Thats a great report GPM, lots of good info.

Can I ask a stupid question, once the autotune has established an AFR that you are happy with over the entire range, do you disable the autotune and just use the maps it has created or is the wideband sensor used all the time so adjustments are being constantly made?

Link to comment
Share on other sites

Thats a great report GPM, lots of good info.

Can I ask a stupid question, once the autotune has established an AFR that you are happy with over the entire range, do you disable the autotune and just use the maps it has created or is the wideband sensor used all the time so adjustments are being constantly made?

Not a stupid question at all!

I think it comes down to personal choice - I guess you could disable it, remove it and then ebay it on to the next chap if you wanted to. However, you can adjust the max amount of variation from the base map that the system will allow (I currently have it set to +10 to -7 from the set cell ammount). As you progress with increasingly more accurate maps (and your trims should start to get smaller and smaller) you can decrease these values (the default range is +20 to -20). That means once you are happy you can largely eliminate the problem on decel simply by not letting the system enrich over a certain amount (less then say +5 is not likely to have a dramatic effect on performance, assuming a good base map). Keeping the Autotune and using the max variation limits also means your bike will continue to refine and adapt as you put different fuels in, temperature changes, even air pressure or altitude. None of these should result in a dramatic change, so tight limits allows your map to remain on target for those AFRs regardless of conditions.

(its things like that which really make the Autotune stand out!)

Link to comment
Share on other sites

Thats a great report GPM, lots of good info.

Can I ask a stupid question, once the autotune has established an AFR that you are happy with over the entire range, do you disable the autotune and just use the maps it has created or is the wideband sensor used all the time so adjustments are being constantly made?

Not a stupid question at all!

I think it comes down to personal choice - I guess you could disable it, remove it and then ebay it on to the next chap if you wanted to. However, you can adjust the max amount of variation from the base map that the system will allow (I currently have it set to +10 to -7 from the set cell ammount). As you progress with increasingly more accurate maps (and your trims should start to get smaller and smaller) you can decrease these values (the default range is +20 to -20). That means once you are happy you can largely eliminate the problem on decel simply by not letting the system enrich over a certain amount (less then say +5 is not likely to have a dramatic effect on performance, assuming a good base map). Keeping the Autotune and using the max variation limits also means your bike will continue to refine and adapt as you put different fuels in, temperature changes, even air pressure or altitude. None of these should result in a dramatic change, so tight limits allows your map to remain on target for those AFRs regardless of conditions.

(its things like that which really make the Autotune stand out!)

Yeah I guess so, I was wondering what the lifespan would be like on the 02 sensors, say for instance it went bad, would that throw everything out or would it revert to the last good map, or would you have to replace it and start over.

I guess like a PC3 you would save a good map to your laptop and then reinstall it if the need arises???

Link to comment
Share on other sites

  • Member Contributer

Yeah I guess so, I was wondering what the lifespan would be like on the 02 sensors

Hundreds of thousands of kilometres!

Ripped from Wikipedia, it says:

"Normally, the lifetime of an unheated sensor is about 30,000 to 50,000 miles (50,000 to 80,000 km). Heated sensor lifetime is typically 100,000 miles (160,000 km)"

The autotune sensors are heated - this is why they must be connected to a 12v power source on the bike.

Link to comment
Share on other sites

Guest Sammy40

Thats a great report GPM, lots of good info.

Can I ask a stupid question, once the autotune has established an AFR that you are happy with over the entire range, do you disable the autotune and just use the maps it has created or is the wideband sensor used all the time so adjustments are being constantly made?

Not a stupid question at all!

I think it comes down to personal choice - I guess you could disable it, remove it and then ebay it on to the next chap if you wanted to. However, you can adjust the max amount of variation from the base map that the system will allow (I currently have it set to +10 to -7 from the set cell ammount). As you progress with increasingly more accurate maps (and your trims should start to get smaller and smaller) you can decrease these values (the default range is +20 to -20). That means once you are happy you can largely eliminate the problem on decel simply by not letting the system enrich over a certain amount (less then say +5 is not likely to have a dramatic effect on performance, assuming a good base map). Keeping the Autotune and using the max variation limits also means your bike will continue to refine and adapt as you put different fuels in, temperature changes, even air pressure or altitude. None of these should result in a dramatic change, so tight limits allows your map to remain on target for those AFRs regardless of conditions.

(its things like that which really make the Autotune stand out!)

Yeah I guess so, I was wondering what the lifespan would be like on the 02 sensors, say for instance it went bad, would that throw everything out or would it revert to the last good map, or would you have to replace it and start over.

I guess like a PC3 you would save a good map to your laptop and then reinstall it if the need arises???

I am liking what i am reading here! Tops efort :fing02: I have just today ordered the PCV. I did not get the auto tune though. Sorry to go :rolleyes: but do you think the auto tune is needed? I will be putting mine on a Dyno as i have NFI how to set this puppy up.

Looks like the PCV is the way to go..<br><br><br>I have been reading now...man what an in-depth analysis of these awesome units!!!<br>

Edited by Sammy40
Link to comment
Share on other sites

  • Member Contributer

Hey GPM, I'm having no luck with my temperature sensor. It never reads above 48 degrees on the Powercommander (I used your starter voltage and temp)

I've tested it with a multimeter, and it's tracking the voltage just fine (and fairly close to the numbers you posted - with a couple of degrees as you say). When the wire is hooked up to the PCV though, it sits on zero until the bike hits 48, then shows 48, but never rises.

Note I am running the stock firmware.

Any advice mate?

Link to comment
Share on other sites

I am liking what i am reading here! Tops efort :fing02: I have just today ordered the PCV. I did not get the auto tune though. Sorry to go :rolleyes: but do you think the auto tune is needed? I will be putting mine on a Dyno as i have NFI how to set this puppy up.

Looks like the PCV is the way to go..

Depends on what you want - people have been using PCIIs and PCIIIs for years without the Autotune option (and normally they opt for a professional map) and I don't think too many of them complain. On the other hand, the Autotune really costs about the same as a good dyno tune anyway.

They way I see it, Autotune is great for people who like to tweak, make progressive changes to their bikes and want to maintain optimum performance, and those who just like fancy gadgets (I fit in all three :D ). If you wanted one I wouldn't advise you against it, but if you don't think it's needed (or don't have the cash) that's a valid option too. The trick with the PCV is you always have the option to add it later if you so desire.

As I posted above the huge advantage of the Autotune is that you are in control of it all - not just relying on a dyno tune to get it right the first time (and they normally do a pretty good job), not relying on tweaking on the fuel map by feel (which is kind of like taking a half educated stab in the dark), but instead you set the end result and the magic box will try to deliver it.

For $250-300 I think it's a good accessory. But hey, I like to tweak, mod and have gadgets wink.gif

Hey GPM, I'm having no luck with my temperature sensor. It never reads above 48 degrees on the Powercommander (I used your starter voltage and temp)

I've tested it with a multimeter, and it's tracking the voltage just fine (and fairly close to the numbers you posted - with a couple of degrees as you say). When the wire is hooked up to the PCV though, it sits on zero until the bike hits 48, then shows 48, but never rises.

Note I am running the stock firmware.

Any advice mate?

First, as it is not clear on what you wrote, you will need to input all the temp voltage points - the PCV does a little magic math to come up with a 'best fit' progression through the points you enter. The more points, the more accurate that is. Obviously if you only enter one voltage point, the PCV will likely treat all voltage readings as that temp.

If that's not the problem, remove the wire from the PCV end and test the voltage there (I assume you were testing at the tap before). The fact you get a reading should mean it's not a continuity error. I had some trouble getting the speed reading to work, turned out to be a bad connection, which is why I removed the pin from the pluged and crimped my tapping wire straight on (problem solved!). All else fails, check all the connections again.

Link to comment
Share on other sites

  • Member Contributer

First, as it is not clear on what you wrote, you will need to input all the temp voltage points - the PCV does a little magic math to come up with a 'best fit' progression through the points you enter. The more points, the more accurate that is. Obviously if you only enter one voltage point, the PCV will likely treat all voltage readings as that temp.

If that's not the problem, remove the wire from the PCV end and test the voltage there (I assume you were testing at the tap before). The fact you get a reading should mean it's not a continuity error. I had some trouble getting the speed reading to work, turned out to be a bad connection, which is why I removed the pin from the pluged and crimped my tapping wire straight on (problem solved!). All else fails, check all the connections again.

Yep, I have all the temp points. I have been measuring the voltage right at the wire that goes to the PCV - ground goes to the battery directly, same as for the PCV.

It's just bizarre.

Link to comment
Share on other sites

Yep, I have all the temp points. I have been measuring the voltage right at the wire that goes to the PCV - ground goes to the battery directly, same as for the PCV.

It's just bizarre.

That's a bit odd. You have them all plugged into the right 'ports'? - from the nearest edge it should be - unused - temp. - speed - rest unused -.

Try removing the wire, grab a jeweller’s screw driver and pock it down there (obviously make sure power is off!) - it may be you didn't pop the membrane fully. Another way to tell you have it in there correctly is when you remove the cable the copper bundle should be squished. If it is still the same shape as when you put it in (likely nice and round) you either are not getting it in far enough or not getting enough contact.

I know the temperature can be a funny beast - going from memory (it's too late to play with my bike right now) you need to set the bike to run (so start the FI sequence, etc) before it will register at all - with the ignition on by kill switch set to 'kill', it should read 160 degrees or so (basically 0 volts) - once the FI sequence is done it will be stupidly inaccurate (as in negative temps) until it starts to approach the first voltage point (so the 48 degrees). As I only use it to tell the PCV when to start listening to the Autotune, I didn't care enough about the oddities to see if they could be fixed (and frankly I don't know if they can be).

I'll try to take a look at my setup tomorrow with a fresh pair of eyes - maybe there is something I have forgotten, seeing as you are getting the right volts at the wire the problem may be software side...

Link to comment
Share on other sites

  • Member Contributer

Great write-up!

I wish I was as good at getting stuff out of my head and onto "paper".

If you want to try the vacuum switch, let me know and I'll hook you up. It makes a big difference if you do a lot of round town riding, not so much if you do mostly highway.

Link to comment
Share on other sites

  • Member Contributer

Holy Power Commander! I wish I hadn't stumbled on to this at lunch. What I read was great. I hope I can follow it when I start tuning!

Link to comment
Share on other sites

Great write-up!

I wish I was as good at getting stuff out of my head and onto "paper".

If you want to try the vacuum switch, let me know and I'll hook you up. It makes a big difference if you do a lot of round town riding, not so much if you do mostly highway.

I wouldn't mind giving the vacuum switch a shot - sadly I have 0 experience with such things so while the theory works in my head I don't know how to translate that into practice (or for that matter, exactly what sort of parts I need to make it all work!). Hence the effort vs frustration tip up.

Link to comment
Share on other sites

  • Member Contributer

That's a bit odd. You have them all plugged into the right 'ports'? - from the nearest edge it should be - unused - temp. - speed - rest unused -.

Try removing the wire, grab a jeweller’s screw driver and pock it down there (obviously make sure power is off!) - it may be you didn't pop the membrane fully. Another way to tell you have it in there correctly is when you remove the cable the copper bundle should be squished. If it is still the same shape as when you put it in (likely nice and round) you either are not getting it in far enough or not getting enough contact.

I know the temperature can be a funny beast - going from memory (it's too late to play with my bike right now) you need to set the bike to run (so start the FI sequence, etc) before it will register at all - with the ignition on by kill switch set to 'kill', it should read 160 degrees or so (basically 0 volts) - once the FI sequence is done it will be stupidly inaccurate (as in negative temps) until it starts to approach the first voltage point (so the 48 degrees). As I only use it to tell the PCV when to start listening to the Autotune, I didn't care enough about the oddities to see if they could be fixed (and frankly I don't know if they can be).

Yeah it's odd alright. I mean it reads *something* because with the bike running when cold it read "0 degrees" and then shows 48 when the bike reaches 48 (as per the table). But after that, it never changes.

Link to comment
Share on other sites

Yeah it's odd alright. I mean it reads *something* because with the bike running when cold it read "0 degrees" and then shows 48 when the bike reaches 48 (as per the table). But after that, it never changes.

I put in a few more pictures at the bottom of my original post which hopefully will help anyone struggling with how they need to hook up the temp and speed wires, but I don't think that's your problem kaldek - after all, you are getting a reading.

I checked through my software and didn't notice anything else you should need to adjust - hook it up, put in the values and off you go. Could you post up screen shots of your temp configuration? Maybe there is some small error in there that is throwing it all out?

Failing that it's back to that universal tool of the amateur electrical tweaker, unhook the relevant wires and redo them...

Link to comment
Share on other sites

I have updated the original post to include graphs (and analysis) of the Autotune with map-by-gear. I wrapped it up in spoiler tags to reduce the scrolling and clutter.

It's all contained within the usage report for the Autotune with map-by-gear enabled wink.gif

Link to comment
Share on other sites

  • Member Contributer

OK - TEMPERATURE READINGS ARE FIXED.

It took upgrading the firmware to 1.4.0 to get it working, however the firmware upgrade was almost a failure.

The firmware upgrade process tells the PCV to go into "boot" mode, which comes up as a device with a different USB ID. Windows XP asked me to reboot to support this new device ID, and I think it caused some form of communication failure. The upgrade process said "error sending file" and the PCV was stuck in boot mode with the green light flashing. I flashed it again, but I had no TPS readings, nothing other than RPM showing.

Anyway I managed to get the firmware installed, but the process was as follows:

  1. Tell the software to upgrade the firmware
  2. Upgrade fails, PCV is stuck in boot mode (green light flashing)
  3. Perform upgrade again
  4. Green light goes solid
  5. Bike starts, but no inputs work. No TPS readings, nothing other than rpms.
  6. Reboot Windows PC to ensure all USB device IDs dynamically detect OK
  7. Perform firmware upgrade again
  8. All problems resolved

During the final firmware upgrade, I noticed that the upgrade process did a lot more stuff - it downloaded not only the firmware file but also a "configuration file" - I would guess this is what made it start working properly again.

Hey GPM, I'm feeling a bit like we're the pioneers mate, forging a path for others to follow once all the gremlins are sorted! :warranty:

Link to comment
Share on other sites

Hey GPM, I'm feeling a bit like we're the pioneers mate, forging a path for others to follow once all the gremlins are sorted! :cheerleader:

Yeah, stick with it boys :biggrin:

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...

Important Information

By using this site, you agree to our Privacy Policy.