Power Commander V w/Autotune Discussion

[ajpagosa]

I wonder if I can reproduce results like this on the SV. I might be able to accomplish that if I force it to run poorly just off idle using the map. Then see how it reacts after the AT turns on. It sure would ease my mind regarding the massive tune changes that keep showing up there.

I downloaded the SV650 maps, interesting thing is they've zero'd the 2% column in the AFR table. And some really big numbers on some of those base maps, including the 2% column. So the AT will never try to adapt there, but the PCV will be adding a lot of fuel. Honestly I am not very impressed with DJ tech people. Not just from this recent interaction either ("exhaust reversion" LOL). It seems the standard answer to "my $$$PCV + AT doesn't run very well" is "turn it off in the region you most need it". And then they make up exotic sounding tech reasons to justify why their product won't work exactly where one would want it to.

Not familiar at all with your model bike but if I were presented with this scenario, I'd add a non-zero value to the 2% AFR column, say 13.2 (with zeros in the 500 and 750 rpm cells like the rest of the map). Select the OTS DJ map that is closest to your mods. Set it up to only turn on after an interval that lets the bike warm up completely to get out of the factory warm up cycle (say 300 seconds) and say +/-20% trim limits. See how that goes.

If you still have problems, take your base map and cut every value in half (cut and paste table into excel), cut trim limits to +/-10%, see how that works.

During cranking and warm up, the ECU on mostly all bikes/cars is running rich and open loop. It has to run open loop because narrow band sensors only work in a narrow region around 14.7:1 AFR, which is the "perfect" ratio for minimal emissions. Once warmed up they go into closed loop mode and target 14.7:1 AFR in the region where the factory has decided to do that for meeting emissions goals. High acceleration regions (rich) or high fuel economy (lean) it will be running open loop because the narrow band sensor cannot measure those AFR's accurately at all, and 14.7 AFR is not good for either. But we can target anything between 11.5 to 16.0 with our AT200 and wideband sensor. The trick in the formerly closed loop region is, part of normal narrow band closed loop ECU operation is constant dithering around 14.7 AFR which still happens if the O2 sensor is removed and it is running open loop, by as much as +/- 0.5 points of AFR (or more nobody knows). So setting up the PCV + AT has to take all that into account, and be fast enough with enough range to cancel it out, but not too much range to get into bad spots.

So you want AT off when the bike is doing its super rich warm up thing because we don't want the AT to try to cancel that out, and on otherwise, everywhere. On top of a base map that isn't too extreme to start with. See if it runs better.

Complete aside here, but consider the logistics of Dynojet offering a PCV + AT for virtually every bike made in the world. Do you really think they've put an example of every one of them on a dyno, run them at every throttle position and rpm possible, found optimal fuel and AFR values for 100 separate cells (or whatever 10x10)? Repeat for every combination of mods? That's like 500 years of dyno time.

I think a lot of these maps and AFR tables are SWAG's based on experience for sure but I'd be stunned if more than 1 in 100 maps and AFR tables were actually gotten off of dyno runs. Or if they were it was a WOT run or two and a maybe a lot of interpolation. I bet DJ has software that takes one dyno run at WOT and some known engine configurations (redline, VE, etc.) and geneartes a bunch of maps for various mods, fills in cells etc. Point is they may be off by a lot. I would not be completely surprised to find out all inline 4 cylinder water cooled maps of all makes and years and mods were extrapolated off of a handful of dyno runs and some theory.

 

[sath182]

I wonder if I can reproduce results like this on the SV. I might be able to accomplish that if I force it to run poorly just off idle using the map. Then see how it reacts after the AT turns on. It sure would ease my mind regarding the massive tune changes that keep showing up there.

I downloaded the SV650 maps, interesting thing is they've zero'd the 2% column in the AFR table. And some really big numbers on some of those base maps, including the 2% column. So the AT will never try to adapt there, but the PCV will be adding a lot of fuel. Honestly I am not very impressed with DJ tech people. Not just from this recent interaction either ("exhaust reversion" LOL). It seems the standard answer to "my $$$PCV + AT doesn't run very well" is "turn it off in the region you most need it". And then they make up exotic sounding tech reasons to justify why their product won't work exactly where one would want it to.

2% is given mostly null values because it doesn't get used at all unless you've got the throttle cracked on deceleration/engine braking. Trying to tune it at anything above maybe 3k RPM (totally guessing here since I can't look at a read out while riding) would be useless. 5 and 10% are likely the cruising points, but 5% is also likely only gonna get up to 4.5-5k RPM before you're cracking into 10%. The bike takes a good bit of throttle to get moving from a start, it doesn't have the bottom end torque to get away without some healthy clutch slipping. That automatically puts you around 2.5k RPM and 10-15% throttle input. Maybe even 20%. So if 2% is used it's in the parking lot right before you down shift.

As for Reversion, it is a thing, but my poking around has shown nearly all examples in relation to to V-8s. If it does happen in a motorcycle, I'm thinking there are two major factors that effect it. I'll wager it's more likely to happen during engine braking, something I love to use. Reason being the vacuum difference in the intake runners with closed throttle plates. And they are right in that this vacuum would have more effect on the exhaust circulation at low RPM when the intake velocity is minimal. I think that will make it real important to avoid negative throttle while trying to tune the bike. That brings me to my other hunch, throttle bores that are too big for lower RPM ranges. Bigger bore, lower velocity for the same engine load. So I can understand reversion happening on a screaming 600 with 38mm TBs serving 150cc chambers each. The SV has 39mm TBs serving 325cc chambers. With both engines running at 3k RPM, the SV will have much higher velocity through the TBs and so less chance of reversion. Which negates the answer I was given by DJ as well.

Not familiar at all with your model bike but if I were presented with this scenario, I'd add a non-zero value to the 2% AFR column, say 13.2 (with zeros in the 500 and 750 rpm cells like the rest of the map). Select the OTS DJ map that is closest to your mods. Set it up to only turn on after an interval that lets the bike warm up completely to get out of the factory warm up cycle (say 300 seconds) and say +/-20% trim limits. See how that goes.

I've already done most of this. The only real difference being that I'm using a custom tune, which I had created a few years ago, that I'm no longer happy with. When I moved here to Augusta, the bike mysteriously started running rich and lost 30 miles per tank. There's a big difference between 185 miles per tank and struggling to get more than 155. This is with no changes to the setup of the bike. I will be lengthening the activation timer though, I think it's set at 120 seconds right now. It's not nearly as cold blooded as the FJR but the extra heat soaking into the heads won't hurt. I'll post up a pic of my current AFR table tonight if I remember to.

If you still have problems, take your base map and cut every value in half (cut and paste table into excel), cut trim limits to +/-10%, see how that works.

I'm not having problems with smoothness, just the unexplained rich condition. Which is my primary reason for reinstalling the AT. I was thinking specifically of chopping up the lower portion of the map to artificially induce a problem and see how the AT works it out.

During cranking and warm up, the ECU on mostly all bikes/cars is running rich and open loop.... So you want AT off when the bike is doing its super rich warm up thing because we don't want the AT to try to cancel that out, and on otherwise, everywhere. On top of a base map that isn't too extreme to start with. See if it runs better.

The SV is a speed density only system. Only the California models from 2007+ have an O2 sensor. Which is also when the twin-spark head was introduced. Also the fast idle period is pretty short. Less than 30 seconds before it settles to its normal idle RPM. It does have a separate set of maps (fuel/ignition/secondary throttle plates) for neutral vs. in gear, though.

Complete aside here, but consider the logistics of Dynojet offering a PCV + AT for virtually every bike made in the world. Do you really think they've put an example of every one of them on a dyno, run them at every throttle position and rpm possible, found optimal fuel and AFR values for 100 separate cells (or whatever 10x10)? Repeat for every combination of mods? That's like 500 years of dyno time.
I think a lot of these maps and AFR tables are SWAG's based on experience for sure but I'd be stunned if more than 1 in 100 maps and AFR tables were actually gotten off of dyno runs. Or if they were it was a WOT run or two and a maybe a lot of interpolation. I bet DJ has software that takes one dyno run at WOT and some known engine configurations (redline, VE, etc.) and geneartes a bunch of maps for various mods, fills in cells etc. Point is they may be off by a lot. I would not be completely surprised to find out all inline 4 cylinder water cooled maps of all makes and years and mods were extrapolated off of a handful of dyno runs and some theory.

You have a point. Though I'll bet the PCV maps aren't too different from the PCiii maps. Especially for a popular bike like the SV that never really saw many changes. With that in mind, there could have actually been some real development on the maps earlier in the SVs life.

 

[eflyguy]

2% is given mostly null values because it doesn't get used at all unless you've got the throttle cracked on deceleration/engine braking. Trying to tune it at anything above maybe 3k RPM (totally guessing here since I can't look at a read out while riding) would be useless. 5 and 10% are likely the cruising points, but 5% is also likely only gonna get up to 4.5-5k RPM before you're cracking into 10%. The bike takes a good bit of throttle to get moving from a start, it doesn't have the bottom end torque to get away without some healthy clutch slipping. That automatically puts you around 2.5k RPM and 10-15% throttle input. Maybe even 20%. So if 2% is used it's in the parking lot right before you down shift.

I disagree, respectfully. In traffic, I often need only the slightest of throttle input and a delicate slip of the clutch to pull away cleanly. That's part of why I love this motor, the torque from idle is more than enough to pull the bike from stationary, or even accelerate in gear.

My Gen II FZ1 is the same way, and requires no throttle whatsoever to get going at the same pace as surrounding traffic, just a smooth release of the clutch and the idle setting is enough.

Both will bog down if you open the throttle more than needed, but will pull smoothly if you open it gradually as the RPMs increase.

 

[Mr_Canoehead]

2% is given mostly null values because it doesn't get used at all unless you've got the throttle cracked on deceleration/engine braking. Trying to tune it at anything above maybe 3k RPM (totally guessing here since I can't look at a read out while riding) would be useless. 5 and 10% are likely the cruising points, but 5% is also likely only gonna get up to 4.5-5k RPM before you're cracking into 10%. The bike takes a good bit of throttle to get moving from a start, it doesn't have the bottom end torque to get away without some healthy clutch slipping. That automatically puts you around 2.5k RPM and 10-15% throttle input. Maybe even 20%. So if 2% is used it's in the parking lot right before you down shift.

I disagree, respectfully. In traffic, I often need only the slightest of throttle input and a delicate slip of the clutch to pull away cleanly. That's part of why I love this motor, the torque from idle is more than enough to pull the bike from stationary, or even accelerate in gear.

I think that Sath182 is talking about his SV-650, not his FJR.

 

[sath182]

2% is given mostly null values because it doesn't get used at all unless you've got the throttle cracked on deceleration/engine braking. Trying to tune it at anything above maybe 3k RPM (totally guessing here since I can't look at a read out while riding) would be useless. 5 and 10% are likely the cruising points, but 5% is also likely only gonna get up to 4.5-5k RPM before you're cracking into 10%. The bike takes a good bit of throttle to get moving from a start, it doesn't have the bottom end torque to get away without some healthy clutch slipping. That automatically puts you around 2.5k RPM and 10-15% throttle input. Maybe even 20%. So if 2% is used it's in the parking lot right before you down shift.

I disagree, respectfully. In traffic, I often need only the slightest of throttle input and a delicate slip of the clutch to pull away cleanly. That's part of why I love this motor, the torque from idle is more than enough to pull the bike from stationary, or even accelerate in gear.

My Gen II FZ1 is the same way, and requires no throttle whatsoever to get going at the same pace as surrounding traffic, just a smooth release of the clutch and the idle setting is enough.

Both will bog down if you open the throttle more than needed, but will pull smoothly if you open it gradually as the RPMs increase.

I'm not talking about the FJR with this statement...

Oh, and here's the AFR table I have loaded right now. The 14.7 AFR in the top left two cells effect the idle. Those weren't there before tonight, but I was watching the readings while finishing up my oil change and noticed it was hitting both cells (yay cell tracer!) and bouncing into the 16s once in a while. The block of 13.6s below is the cruising range all the way up to 75mph.

I'm keeping an eye on the 2% and 5% columns as well as the lower RPMs of the 10-20% columns because they are getting some big numbers in them. The 2% and 5% ranges I think I'll end up resetting to the custom map values and leaving them alone. But 10% at 1750rpm is now sitting at +35 and the one next to it is +25. My brother just mentioned that this might be necessary as a sort of accelerator pump. I wonder just how fat that is. Really, the whole lower corner of the thing will probably get reset to the start and the target AFR turned off. I'll probably leave the cruising zone, the areas surrounding it and the rest of the map turned on with an allowable correction delta of 5% or less.

The current map.

The starting map. Compare the cruising block and you'll notice its now quite a bit leaner in the current map

 

[eflyguy]

Sorry, I missed that.. I'll go sit in the corner again..

 

[ajpagosa]

Sath182,

What exactly is the problem you're having with your SV? I see you do have some values in the 2% column on your AFR table.

Very interesting that they made a 2009 model with EFI but without an O2 sensor for 49 states, and then one with for CA. It may be they just took the O2 sensor off the CA bike and left its ECU as a default open loop or reprogrammed it for no sensor. Most of my comments about setting up an AT still apply. An open loop non-O2 sensor speed density system still has a warm up cycle and you'll want AT off for that. It almost certainly has its own type of internal trim tables which it will be constantly adjusting. For all that and to offset any other unknown weirdness the ECU is doing, if that in fact is the source of your problems, you'll still want a closed loop wideband on top of it all. A closed loop wideband system will solve a lot of fueling problems due to odd ECU behavior but not all of them, something else could be going on outside the scope of fuel management. The trick with the FJR was to figure out how to set up the AT to counteract those specific issues but if its problems were due to a bad sensor, vac line leak, or faulty fuel pump for example (or many other things) it would not help much.

Back to your SV, there is an accel pump utility on the PCV. Set up right it works well, set up wrong it is awful. I have messed around with it on the FJR and several other bikes. On the FJR I've just turned it off, wanting to isolate only the wideband + base map and concentrate on it. On 2 other bikes with PCV + AT (Roadliner and WR250R) it works pretty well smoothing out throttle transitions. You have to activate it up in one of those pull down menus, there's some stuff out there on the web explaining the settings. I would not try to use the basic PCV fuel map to simulate an accel pump.

The 08 WR250R and the 06 Roadliner both have many mods, worked well with PCV + AT right out of the box. Accel pump is running on them and improved operation. FJR, not so much, and had all these other issues.

 

[sath182]

Sorry, I missed that.. I'll go sit in the corner again..

Sath182,What exactly is the problem you're having with your SV? I see you do have some values in the 2% column on your AFR table.

A mysterious drop in tank range is the problem I'm trying to solve. the whole time I've owned the bike it got 186 miles on a tank before hitting reserve. After this past move, it getting a miserable 155 per tank. The target AFR values in the 2% column I added with my previous round of tinkering so I could see what it would do. I'm not liking the way it's changing things so I'm going to blank those and some other areas out. I'm happy so far with where the cruising range has settled, but I need a day on the bike with the AT turned off to see if it's helped the tank range.
I've literally been through the bike top to bottom this last winter. (full engine tear down to trouble shoot a transmission engagement issue) There are no fuel leaks, the injectors are free and clear, the plugs are only a year old, the air filter is clean, and there are no exhaust leaks. It's really down to just running rich, which I can smell at a stop and couldn't before moving. Oh, and E-10 vs. E-Free makes no difference.

Very interesting that they made a 2009 model with EFI but without an O2 sensor for 49 states, and then one with for CA. It may be they just took the O2 sensor off the CA bike and left its ECU as a default open loop or reprogrammed it for no sensor. Most of my comments about setting up an AT still apply. An open loop non-O2 sensor speed density system still has a warm up cycle and you'll want AT off for that. It almost certainly has its own type of internal trim tables which it will be constantly adjusting. For all that and to offset any other unknown weirdness the ECU is doing, if that in fact is the source of your problems, you'll still want a closed loop wideband on top of it all. A closed loop wideband system will solve a lot of fueling problems due to odd ECU behavior but not all of them, something else could be going on outside the scope of fuel management. The trick with the FJR was to figure out how to set up the AT to counteract those specific issues but if its problems were due to a bad sensor, vac line leak, or faulty fuel pump for example (or many other things) it would not help much.

The SV was carbureted when it came out in 99. In 03 it was updated with the speed density EFI, and it's been largely the same since. The only changes in 07 were twin spark heads, changing the fast idle cam assembly and idle adjustment screw for a computer actuated idle control valve, updated fuel maps, and the O2 sensor on the CA models.

Back to your SV, there is an accel pump utility on the PCV. Set up right it works well, set up wrong it is awful. I have messed around with it on the FJR and several other bikes. On the FJR I've just turned it off, wanting to isolate only the wideband + base map and concentrate on it. On 2 other bikes with PCV + AT (Roadliner and WR250R) it works pretty well smoothing out throttle transitions. You have to activate it up in one of those pull down menus, there's some stuff out there on the web explaining the settings. I would not try to use the basic PCV fuel map to simulate an accel pump.
The 08 WR250R and the 06 Roadliner both have many mods, worked well with PCV + AT right out of the box. Accel pump is running on them and improved operation. FJR, not so much, and had all these other issues.

I don't really accept it as an explanation, considering the accel pump-ness would be according to any rpm/throttle combo you demand when making a change. Especially big ones. So the +35 and +25 cells are probably going to go back to +18 where they were in the custom map. If I keep a target AFR in that area, I'll probably set it higher. This is the part of the map where the engine bogs to when letting the clutch out while starting from a stop. I've never had any problems with performance there, and generally it always seemed pretty well sorted low down.
Question about the built in accel pump though... are the transitions you're referencing like those you make when getting back on the throttle mid-corner to exit and want it to be nice and smooth?

 

[ajpagosa]

Question about the built in accel pump though... are the transitions you're referencing like those you make when getting back on the throttle mid-corner to exit and want it to be nice and smooth?

Yes. It richens up mixture briefly on throttle transitions. You can set it by how much and how many crank revolutions of enrichment, how much throttle threshold etc. However if you are smelling gas and bogging like it is too rich already off idle and other throttle/load transitions, you may want to lean out the base map before you use it. These were similar to problems I had with the FJR, it would bog and generally behave as if it were running to rich even though it was targeting AFR's that weren't IMHO. I think the base DJ OTS map was very aggressive for my altitude, that plus the wide AT adjustment and other ECU issues would get it into a spot where it was drowning in fuel and the AT couldn;t get out of the hole. So adding more fuel in my case with accel pump was not productive. Whatever the reason on the FJR, I wanted to debug the AT and base map only so turned it off.

On the other two bikes where I am running it, it was to fix a jerky throttle and slight bog right when you crack throttle off cruise and idle somewhat independent of rpm. It's like they would wait a beat, stumble slightly then take off. Not a huge problem but annoying. Accel pump cleared that up, and it was running with PCV + AT on. These bikes never got into the hole the FJR did, like getting progressive worse until its unridable. It was more like fine tuning some minor things.

 

[sath182]

I'm not having any performance problems at all. The bog I mentioned in relation to the +35 and +25 cells is just from the clutch sometimes dragging it down there on a start.

The second part of your reply answered my question quite directly though. Thanks for that. Sounds like I probably won't need to fool with the feature as I can be pretty smooth on the SV without it. That's totally down to having a properly trained right hand though.

 

[sath182]

I got out for a 400+ mile day yesterday and can report success with the tuning in my cruise range. The first tank came out to 54 mpg and the second was 56 mpg. Right now it has 164 on the trip meter for the third tank and I'm gonna see how far it goes till the light comes on. I think I'm pretty well done futzing with this map, I'll probably review the trims and save it off as a finished product very soon. I'm not done though I'm gonna take the target AFR table for the FJR and re-work the SV map to use the leaner values in the FJR map. See if I can tweak over 60 mpg out of this thing. First thing's first though... I need to check my coolant level, my engine temps were all over the place yesterday.

 

[Mr_Canoehead]

If your coolant level is good, your thermostat is probably going out. If it is running cold, your mileage would definitely take a hit.

 

[sath182]

It's running hot actually. Historically it's always run around 168F unless the ambient temperature is in the 100s. Yesterday's return trip saw it staying around 175 only when running 60+ with no traffic in front. If we caught up to a car, the temps would start climbing to around 185 until we passed the car and got back into clean, cool air. Hitting a stop light or 35 mph zone let it get up to 200 very quick. There's no froth in the oil, no smoke coming from the exhaust, no sweet smell of freshly leaked coolant, and the water pump weep hole is dry. I'm hoping I just didn't get an effective bleed of the cooling system that is only showing itself now since I'd only been riding it to and from work until yesterday.