idle & TPS question - another ECU

Post Reply
sparky
Posts: 21
Joined: Tue May 22, 2018 10:24 pm

idle & TPS question - another ECU

Post by sparky »

So a couple of buddies and I have aftermarket ECUs on our dirt race race bikes. They are 2 stroke EFI. They are from a company called GET.

So here is the funny thing. The TPS the actual raw values are like 450 (throttle closed) to 3500 (throttle open). So when we calibrate the ECU to the "actual" TPS values, the idle has a tendency to "hang", the AFR meter shows lean (like 15 or 16).

So, the manual for the software says to calibrate the TPS to +15 from the actual on fully closed, and -15 from the actual on fully open. OK. So we set the TPS to 465 on closed and 3485 on fully open. The idle still has a tendency to "hang", the AFR meter shows lean (like 15 or 16).

OK, fine, so we try the opposite: Set the TPS to 425 and 3525 - in other words lower and higher than the actual TPS. So doing this: presto! the idle drops nicely and the AFR shows nicely around 12.8 or so...

Thinking this through, we have deceived the ECU to think that when the actual throttle is closed, the ECU thinks it is partially open and is revving a little bit - in other words the injection/ignition is reading from the map, where we specifically set fuel quantity.

Unfortunately GET uses proprietary mapping software, and has nothing for idle fuel mixture. So we have no idea what is really going on. *sigh*

Any thoughts as to what might be happening?
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: idle & TPS question - another ECU

Post by kb1gtt »

What do you mean hang? Are you saying the idle is a higher RPM than desired? Perhaps the air is limiting the RPM, but it's dumping more fuel, or sparking later which is causing the AFR to show wrong. I'm not following what you mean by hang.

How are you doing your AFR's, as O2 sensors would likely get buggered fairly quickly with the oil in the gas.

Perhaps there is some idle code which is activated when your TPS is close to the min, perhaps there is simply a bug. I don't know. It sounds like the configuration software lacks configuration options for when your TPS is closed. Perhaps they developed some code there and planned that future upgrades could open new features. Who knows. It sounds like it could be many things. However it also sounds like you have found that if you tell the ECU it's not full closed, you can then tune it via tables.

If it works how you want it when the ECU thinks the TPS is slightly open, instead of closed, then I would say run with it. I think that also aligns with typical small engine mechanical design. Emissions regulated engines tend to have Idle Air Control (IAC) valves which will regulate the air to keep a smooth idle. To know when this valve is needed, it's common they detect that via the TPS closed position. This allows a smooth idle at sea level or at the top of a mountain. In your case a dirt bike doesn't generally see that kind of variation in ambient pressures so they often negate it as it's just extra weight and doesn't offer features that the operator desires.
Welcome to the friendlier side of internet crazy :)
sparky
Posts: 21
Joined: Tue May 22, 2018 10:24 pm

Re: idle & TPS question - another ECU

Post by sparky »

"Hang" or "hanging idle" is a common term with bikes and sleds. I guess not here. It means when letting off the throttle the revs do not come down to idle or come down slowly.

The O2 sensors we use (with adapters) work extremely well for two stroke tuning. Placement is key within the expansion chamber. We also have very low mix ratios (100:1 to 120:1). These have been used for a lot of years for 2 stroke tuning on sleds and bikes and with dyno tuning. We are very confident with the numbers.

I am just looking for a possible explanation of the ECU behavior with TPS and idle.
sparky
Posts: 21
Joined: Tue May 22, 2018 10:24 pm

Re: idle & TPS question - another ECU

Post by sparky »

Hanging idle is almost always a lean condition.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: idle & TPS question - another ECU

Post by kb1gtt »

Hmmm, I recall what might be considered hanging with a carburetor bike. Would you expect hanging on a carburetor bike? I recall that when in neutral and idle, it would take a long time for the engine RPM to come down. I just figured it was inertia, but now that you mention it, I have only noticed it on bikes, not on chain saws, go carts, or other smaller 2 stroke applications. I guess I'm ignorant about idle hanging. Perhaps it's an XTau situation, were the wet wall keep delivering fuel even though the injector has stopped injecting.

Do you happen to know of a youtube or similar about hang? If I see the situation I might be able to offer more comments.

Are you using a LSU4.9 for you WO2? I'm curious what O2 sensor you are saying is working well. I've not see any 2 cycles with an O2. It's good to hear you feel confident in it's numbers.
Welcome to the friendlier side of internet crazy :)
sparky
Posts: 21
Joined: Tue May 22, 2018 10:24 pm

Re: idle & TPS question - another ECU

Post by sparky »

Hanging idle is a very common occurrence on bikes with carbs that are lean in the pilot, or needle circuits. It is a trick we use often for tuning carbs on bikes.

The o2 sensors are only used for tuning, then removed. Their life on a two stroke is not long enough for permanent mount. Check out the sled guys, they are big into EFI and tuning using o2 sensors. I use a 4.9, other use 4.2. In both cases a special adapter is used to reduce oil contamination. Mounting location and position are key, as well as orientation of the special adapter. A lot of the times we will use relative comparisons rather than absolute comparison for tuning, but it works, and works well.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: idle & TPS question - another ECU

Post by kb1gtt »

From what I can tell, the "lean" condition means the side walls are wet, and even though the carb or injection have stopped sending fuel, there is enough on the side walls to keep the engine rev'ed up for a while when unloaded. From experience and from my brief look at the interwebs, it seems this is only when in neutral or when the clutch has separated the engine from the drive chain. When driving you don't notice a significant amount of power generated by this. However when shifting gears, the syncro's really want the RPMs to be normal.

With carb's, it's a bugger as you cannot make XTau type corrections, and you need to deal with wall wetting mechanically. If you change your jets, get an air leak, or do other mechanical things, it can effect the wall wetting.

However with EFI you have more options to compensate for the wet wall(s). It sounds like your ECU simply doesn't have that option, and you don't have the ability to add it.

By making this controlled via tuning table at idle, I think you been able to find a semi-happy medium between the wall wetting and long term idle. I would guess one of the below.
-- Long term idle is lean. After say like 1 minute of idle you'll find your AFR is low. This can be risky as idle can cause heat build up, which can cause problems with detonation or cyl erosion.
-- You may not have a good idle after several seconds. Perhaps even stall, but only after is has run for like 30 seconds to a minute or something similar.

If you have a lean idle, I'd avoid idling for prolonged periods of time. If you have a capable EFI, you could compensate for these conditions and get a good idle, while preventing the idle hang. However it sounds like the system you have, you should simply avoid sitting around idling.
Welcome to the friendlier side of internet crazy :)
sparky
Posts: 21
Joined: Tue May 22, 2018 10:24 pm

Re: idle & TPS question - another ECU

Post by sparky »

Handing idle is not like what you read of the internet. It happens not only at idle but when the bike is running and engaged. You have to experience it to understand.

So lets skip the hanging idle thing and I will rephrase my question: is it common for ECUs have a separate idle fuel map that is outside of the main fuel map?
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: idle & TPS question - another ECU

Post by kb1gtt »

I have not seen a separate table for idle. It might exist but I have not heard or seen of such a thing. What I have seen is algorithm's like XTau which deal's with many issues including wall wetting. I've also seen code that runs an IAC valve to help address air flow issues at idle. Things like that do commonly exist and are commonly used to increase idle performance or to help with dynamic throttle issues. I have also seen temperature tables, which make compensation's based on a temperature look up table. But I have not see a separate MAP specifically for idle.

As for if I see a potential value in a separate table for idle conditions. I can't really think of much in way of benefits. Perhaps such a table could allow for more data points, and a more refined tuning at idle conditions. AKA when at high RPM's with WOT if your load is 90% vs 100%, there's not a huge change in the fuel command. However at low RPM if you change or load from 10% to 20%, you have a much larger change in fuel command. If your primary fuel table was limited by memory, lets pretend it was something crazy low like a 4X4 table, I could see how this could cause issues with the bottom left area of the table. So if your MCU was limited like this, perhaps they simply switched to a different table when in that region to gain increased accuracy at that time. AKA if you had another 4X4 table that was used when below say 1kRPM and say below 50% load, a second table in that situation could greatly help with the precision in the area that needs additional precision. I have not seen this, but that is a possible benefit of a second table. Other than that I can't think of much use in a second fuel table for idle conditions.
Welcome to the friendlier side of internet crazy :)
mck1117
running engine in first post
running engine in first post
Posts: 1493
Joined: Mon Jan 30, 2017 2:05 am
Location: Seattle-ish

Re: idle & TPS question - another ECU

Post by mck1117 »

kb1gtt wrote:
Thu Dec 06, 2018 10:08 am
As for if I see a potential value in a separate table for idle conditions. I can't really think of much in way of benefits. Perhaps such a table could allow for more data points, and a more refined tuning at idle conditions.

...

Other than that I can't think of much use in a second fuel table for idle conditions.
The use I've seen for a separate idle fuel table is on engines with an idle control valve (as opposed to electronic throttle, and no other flow path). Throttle blade vs. IAC valve may have very different VE. Even though the manifold pressure might be the same, how the air got there could change the VE considerably. It matters if you want bang on idle/light throttle emissions, but for most people it /probably/ doesn't matter much.
Post Reply