Some thoughts on PCV/Autotune tuning...

[sEvOn]

It seems a number of people are having issues with tuning the Power Commander and the Autotune add-on for cruise/part throttle.

In the interest of seeing people spend less time and money on tuning, and more time riding, here's my take:

The Power Commander can give you good results for 3/4 to full throttle but is going to be frustrating/inconsistent for part throttle.

The factory Husky fuel injection uses a manifold pressure sensor (MAP), throttle position sensor (TPS), coolant temp sensor, air temp sensor, and RPM to calculate fueling (known as speed density). The Power Commander overlays this stock fuel map with one that uses only the TPS signal and RPM (known as alpha-n).

Using a MAP sensor (speed density) is a much more accurate way of tuning fuel injection because it takes into account engine load by using the MAP sensor signal, along with the throttle position. Alpha N systems are really only accurate at idle and large throttle opening, everything in-between, like part throttle, is a guess at what engine load really is. For a given throttle position, load can change due to what gear you're in or whether you're going up/downhill.

Some fuel injection systems transition from speed density to alpha-n after a certain load/throttle position. Not sure if the Husky system does this, it may though. This is because the engineers know the alpha-n system can be accurate/useful at large throttle positions, but not-so-much at part throttle.

The autotune add-on attempts to counteract the fact that alpha-n isn't accurate at part throttle, but this introduces an issue for us single cylinder folks. The exhaust pulses from a single cylinder can provide inaccurate wideband 02 readings at idle/low rpm and part throttle. At low RPM there's gaps between the exhaust pulses coming from the engine and the sensor expects a constant exhaust flow (fairly constant atleast). Part throttle situations also dont have alot of exhaust flow, and when combined with gaps between pulses at low rpm you get a signal that may not be accurate. Try taking a wideband reading at idle with a high sample rate and you'll see what I mean.

As mentioned, the alpha-n system works well at/near WOT. And at higher RPM and higher engine loads, the wideband signal becomes much more accurate/reliable. If you're tuning for that situation then the Power Commander will work as promised, but for part throttle/midrange on single cylinder bikes you're likely to have a tough time.

Disclaimer: I've got some experience with standalone FI systems for cars (both speed density and alpha-n) and have tuned single cylinder go-kart engines in the past. Both are hobbies. Just wanting to share some things I've learned along the way.

 

[moto66]

Thanks for sharing some good information

 

[wilmar13]

The factory Husky fuel injection uses a manifold pressure sensor (MAP), throttle position sensor (TPS), coolant temp sensor, air temp sensor, and RPM to calculate fueling (known as speed density). The Power Commander overlays this stock fuel map with one that uses only the TPS signal and RPM (known as alpha-n).

Isn't PC a piggy back that is only augmenting the stock EFI? In other words it isn't overriding it, just tweaking it with the MAP, IAT, etc still in play.

The autotune add-on attempts to counteract the fact that alpha-n isn't accurate at part throttle, but this introduces an issue for us single cylinder folks. The exhaust pulses from a single cylinder can provide inaccurate wideband 02 readings at idle/low rpm and part throttle. At low RPM there's gaps between the exhaust pulses coming from the engine and the sensor expects a constant exhaust flow (fairly constant atleast). Part throttle situations also dont have alot of exhaust flow, and when combined with gaps between pulses at low rpm you get a signal that may not be accurate. Try taking a wideband reading at idle with a high sample rate and you'll see what I mean.

Yep, I can verify that! And the single cylinder aspect makes more sense too to exacerbate the issue. Dynojet says to leave the target AFR for 0% and 2% throttle position at "0" (autotune won't be active) but I found you have to do this up to 10% TP or else it leans the bike out. This setup seems to have no downside. It is smooth everywhere and runs well in all conditions. With the stock EFI it would not run well, especially at altitude and great changes in humidity. Not sure if the MAP and IAT signals are poor or what, but the PCV with autotune has transformed the bike in terms of rideability and throttle response at all loads and under all conditions (you have to disable autotune below 10% TP though!)

 

[Tee-Dub]

I'm finding this with mine too. Hard on the gas and its a beast, but cruising around it keeps getting lean and feels like crap. I'm about ready to disable it for the 10% column too and manually enter numbers for that range. I've even debated on removing the auto tune and having a custom dyno map generated, but I'm not ready to give up yet!

 

[husky bom]

removing auto tune and getting a custom dyno map is the best thing you could do for your bike
auto tune is only good collecting data

 

[moto66]

removing auto tune and getting a custom dyno map is the best thing you could do for your bike
auto tune is only good collecting data

I've got my 09 SM510R booked in Saturday this week for a PCV custom dyno map. No autotune. I'll ask for a before and after graph and post it up with the mods I've done.

 

[wilmar13]

removing auto tune and getting a custom dyno map is the best thing you could do for your bike
auto tune is only good collecting data

I've got my 09 SM510R booked in Saturday this week for a PCV custom dyno map. No autotune. I'll ask for a before and after graph and post it up with the mods I've done.

My Aprilia was dyno tuned with PC3 and I have a nice before after map of AFR, with the before showing rich at low throttle and after showing “perfect 13:1” across the rpm range… it had the same problem the Husky did with Autotune, lean as hell at idle and low throttle settings… but that couldn’t be right, since it was tuned at a dyno!… or could it be that errors in measuring AFR’s are related to the method (wideband O2 in exhaust stream) not the equipment?

I deleted all the negative fuel trims in the PC3 map the “dyno expert” put in the low rpm and TP’s and the bike runs great now, same as the PCV with AT on my Husky.

My only point is that the flaw of Autotune is the same on a dyno. If you think you can avoid that, you are kidding yourself. If you don’t want to mess with it, and trust that the dyno guy won’t jack with your mapping except under load where there is sufficient exhaust flow to get accurate O2 sensor readings then that is a good way to go(and cheaper)… but don’t think that Autotune is inferior to a dyno. They are the same method… one you have control over and can do it over and over under many different conditions, one you have no control over and can only be done under one condition.

For accurate measurement of power changes due to mods, the dyno is superior since you have most variables fixed (i.e. control environment). For power optimization through fuel mapping (the point of PCV in the first place) Autotune is superior since you measuring it with lots of variables and under many conditions (i.e. the real world). That is an important distinction.*

*- assuming Autotune works as intended…

 

[moto66]

My Aprilia was dyno tuned with PC3 and I have a nice before after map of AFR, with the before showing rich at low throttle and after showing “perfect 13:1” across the rpm range… it had the same problem the Husky did with Autotune, lean as hell at idle and low throttle settings… but that couldn’t be right, since it was tuned at a dyno!… or could it be that errors in measuring AFR’s are related to the method (wideband O2 in exhaust stream) not the equipment?
I deleted all the negative fuel trims in the PC3 map the “dyno expert” put in the low rpm and TP’s and the bike runs great now, same as the PCV with AT on my Husky.

My only point is that the flaw of Autotune is the same on a dyno. If you think you can avoid that, you are kidding yourself. If you don’t want to mess with it, and trust that the dyno guy won’t jack with your mapping except under load where there is sufficient exhaust flow to get accurate O2 sensor readings then that is a good way to go(and cheaper)… but don’t think that Autotune is inferior to a dyno. They are the same method… one you have control over and can do it over and over under many different conditions, one you have no control over and can only be done under one condition.

For accurate measurement of power changes due to mods, the dyno is superior since you have most variables fixed (i.e. control environment). For power optimization through fuel mapping (the point of PCV in the first place) Autotune is superior since you measuring it with lots of variables and under many conditions (i.e. the real world). That is an important distinction.*


*- assuming Autotune works as intended…

Good info Wilmar

I'm at dead sea level and only pleasure riding, 90% street use, dry etc so am hoping the custom map will fit the bill for my needs. I've always dynoed sports bikes and have always been happy for my use. I don't like the cost though!

 

[wilmar13]

Good info Wilmar I'm at dead sea level and only pleasure riding, 90% street use, dry etc so am hoping the custom map will fit the bill for my needs. I've always dynoed sports bikes and have always been happy for my use. I don't like the cost though!

I think Autotune only makes sense if you like playing with stuff, you change mods more than once or twice, or you have multiple bikes and will use PCV on others.

It costs about 3x a dyno tune. The real question is not whether it will save you 3 trips to the dyno, but do you want to have control over your bike or let someone else do it. One isn’t better than the other, I just wanted to point out that the bad values under no/low load are inherent with both dyno and autotune since they both use a wideband O2 to indirectly measure A/F in the exhaust stream and then compensate based on those reported values.

If you don't want to mess with it, taking it to someone is probably the way to go and you should end up with the same result for less moohla.

 

[moto66]

I think Autotune only makes sense if you like playing with stuff, you change mods more than once or twice, or you have multiple bikes and will use PCV on others.

It costs about 3x a dyno tune. The real question is not whether it will save you 3 trips to the dyno, but do you want to have control over your bike or let someone else do it. One isn’t better than the other, I just wanted to point out that the bad values under no/low load are inherent with both dyno and autotune since they both use a wideband O2 to indirectly measure A/F in the exhaust stream and then compensate based on those reported values.

If you don't want to mess with it, taking it to someone is probably the way to go and you should end up with the same result for less moohla.

I definitely don't want to be tweaking with the fuelling once it's set. Doesn't a PCV set up do away with the 02 sensor? Mine went with the Husky power up kit fitting. Thought the 02 is needed if you do run the autotune? Or do you mean the AFR sensor they stick in the exhaust during the dyno tune? Good thing is my local dyno shop is only 2 miles away so any tweaks it's £45.00 a run which is a lot less than AT.

 

[Tee-Dub]

I paid $219 for my Auto Tune. I've had custom dyno maps made for other bikes and they weren't a lot cheaper, let alone 1/3 the price of the auto tune. I think EDR in Portland charged me about $150 or maybe a little more for the last dyno map they did for my Aprilia Tuono and that was 4 years ago.