Page 2 of 6

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 12:29 pm
by ptuomov
peejay wrote: Sun May 13, 2018 10:39 pm
joe 90 wrote: Sun May 13, 2018 5:28 pm Lean spool? There's more than one way to do it. If you run extra rich (11s) when you don't need to , you kill the power output.
So keep it lean, normal lean, in the 12s or 13s......
Modern turbocars run stoich in boost, sometimes up to 15-17psi.

I didn't really believe it either until I saw the numbers. There is a LOT of attention paid to cooling system dynamics. Ford has engines with three-zone cooling systems that can focus on specific parts of the head!
I am interested in producing most WOT torque at each rpm below the rpm point where the wastage opens under steady state load. I don't particularly care about fuel consumption or emissions in WOT operation, it just needs to go fast. At those low rpms, the total amount of heat that needs to be shed from the engine isn't that large compared to full boost at redline, so I don't anticipate cooling to be an issue. Also, it may not be particularly knock limited at those rpms either, as the turbos haven't spooled and ignition retard may indirectly increase torque via higher turbine speed and higher boost.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 2:43 pm
by ijames
ptuomov wrote: Mon May 14, 2018 12:29 pm I am interested in producing most WOT torque at each rpm below the rpm point where the wastage opens under steady state load. I don't particularly care about fuel consumption or emissions in WOT operation, it just needs to go fast. At those low rpms, the total amount of heat that needs to be shed from the engine isn't that large compared to full boost at redline, so I don't anticipate cooling to be an issue. Also, it may not be particularly knock limited at those rpms either, as the turbos haven't spooled and ignition retard may indirectly increase torque via higher turbine speed and higher boost.
The total cooling system heat load at lower rpms may not be that much but the hotter you get the combustion chamber and piston and exhaust valve at lower rpms the hotter they will be later when you are approaching peak hp, which will lower your detonation resistance and probably lower the peak safe hp. I think it will be a balancing act - add heat for spooling until you start to see it affect the peak hp. I'm no guru but I have spent a good bit of time making chips for the stock ecm in the 86-87 Buick turbo V6's, and I spent a lot of time tuning the spoolup fueling. I would start logging at idle, mat the pedal, and lift when the boost was over 6-7 psi. Then go back to the log and measure how much time elapsed between the TPS hit and an arbitrary MAF value that was in the 3-4 psi range, and use this time as my spoolup score for that chip. Then look over which timing and fuel enrichment cells were hit, look at the narrow band (and later wideband) O2 values, make some changes, burn a new chip, and do it all over and over again. I also did this starting by brake torquing up against the torque converter to a fixed rpm and boost, and then matting it. Luckily I found that both techniques gave basically the same fuel curve, and that wound up just a bit leaner than the best WOT values at higher rpms. Maybe 12 to 12.5 for spoolup and 11 to 11.5 at WOT above 4000 rpm at 22-23 psi in 3rd gear with 93 octane pump gas. Going leaner did not help spoolup, and if it got lean enough to start detonating during spoolup that detonation would continue into the higher boost and rpm ranges even though the fuel map up there was perfectly fine if it did not start detonating at lower rpms. That's why I say you need to worry about too much heat in the chamber too soon, that will limit you long before the cooling system will.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 2:53 pm
by ptuomov
ijames wrote: Mon May 14, 2018 2:43 pm
ptuomov wrote: Mon May 14, 2018 12:29 pm I am interested in producing most WOT torque at each rpm below the rpm point where the wastage opens under steady state load. I don't particularly care about fuel consumption or emissions in WOT operation, it just needs to go fast. At those low rpms, the total amount of heat that needs to be shed from the engine isn't that large compared to full boost at redline, so I don't anticipate cooling to be an issue. Also, it may not be particularly knock limited at those rpms either, as the turbos haven't spooled and ignition retard may indirectly increase torque via higher turbine speed and higher boost.
The total cooling system heat load at lower rpms may not be that much but the hotter you get the combustion chamber and piston and exhaust valve at lower rpms the hotter they will be later when you are approaching peak hp, which will lower your detonation resistance and probably lower the peak safe hp. I think it will be a balancing act - add heat for spooling until you start to see it affect the peak hp.

I'm no guru but I have spent a good bit of time making chips for the stock ecm in the 86-87 Buick turbo V6's, and I spent a lot of time tuning the spoolup fueling. I would start logging at idle, mat the pedal, and lift when the boost was over 6-7 psi. Then go back to the log and measure how much time elapsed between the TPS hit and an arbitrary MAF value that was in the 3-4 psi range, and use this time as my spoolup score for that chip. Then look over which timing and fuel enrichment cells were hit, look at the narrow band (and later wideband) O2 values, make some changes, burn a new chip, and do it all over and over again. I also did this starting by brake torquing up against the torque converter to a fixed rpm and boost, and then matting it. Luckily I found that both techniques gave basically the same fuel curve, and that wound up just a bit leaner than the best WOT values at higher rpms. Maybe 12 to 12.5 for spoolup and 11 to 11.5 at WOT above 4000 rpm at 22-23 psi in 3rd gear with 93 octane pump gas. Going leaner did not help spoolup, and if it got lean enough to start detonating during spoolup that detonation would continue into the higher boost and rpm ranges even though the fuel map up there was perfectly fine if it did not start detonating at lower rpms. That's why I say you need to worry about too much heat in the chamber too soon, that will limit you long before the cooling system will.
interesting method and data on optimal AFRs. It seems like you just ended up with the AFRs that would give about the best torque in a normally aspirated engine.

I agree that if you let it start detonating, you'll have to cut boost and ignition advance more to end detonation that started in previous cycles than what would be required for detonation to start without previously detonating cycles. However, if it is not detonating, my logic tells me that heat management at 2500 rpm when the engine is making under 200 hp is a lot easier than heat management at 6000 rpm when it's making four times that. This even considering that the mass flow thru the engine is a lot lower.

Did you find that the car accelerated the fastest below boost threshold rpm with the ignition advance near the knock limit or with the ignition significantly retarded from that? Later ignition timing helping the turbine to some extent.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 3:56 pm
by bentvalves
[/quote]Modern turbocars run stoich in boost, sometimes up to 15-17psi.

I didn't really believe it either until I saw the numbers. There is a LOT of attention paid to cooling system dynamics. Ford has engines with three-zone cooling systems that can focus on specific parts of the head![/quote]


I believe this statement to be false sir. What OEM vehicle boosts through peak torque at 14:7:1?

Generally the OEM's are are pig rich in this region of the fuel map. (11's)

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 4:42 pm
by ptuomov
This is the closest they've gotten so far with the AFRs in gasoline turbo engines:

Image
Image: Ecoboost 3-cylinder direct injection gasoline engine (lambda map)
Credit: Ford

For example, the Ford Ecoboost 3-cylinder engine runs with stoichiometric air-fuel ratio for idle to medium engine speed and complete load range, and with rich air-fuel mixture at high speed and load. The reason for which it runs with rich mixture at high engine speed and load is engine cooling. The additional fuel (which will remain unburnt) is injected to absorb heat (through evaporation), reducing this way the temperature in the combustion chamber.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 5:46 pm
by bentvalves
is the above engine direct-injected?

If so, it is a completely different animal than a port-injected engine.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 6:59 pm
by ptuomov
bentvalves wrote: Mon May 14, 2018 5:46 pm is the above engine direct-injected?

If so, it is a completely different animal than a port-injected engine.
Yes it is, direct injected. I don’t know if it runs as homogenous charge or stratified charge engine at WOT under boost.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 10:00 pm
by peejay
ijames wrote: Mon May 14, 2018 2:43 pm
ptuomov wrote: Mon May 14, 2018 12:29 pm I am interested in producing most WOT torque at each rpm below the rpm point where the wastage opens under steady state load. I don't particularly care about fuel consumption or emissions in WOT operation, it just needs to go fast. At those low rpms, the total amount of heat that needs to be shed from the engine isn't that large compared to full boost at redline, so I don't anticipate cooling to be an issue. Also, it may not be particularly knock limited at those rpms either, as the turbos haven't spooled and ignition retard may indirectly increase torque via higher turbine speed and higher boost.
The total cooling system heat load at lower rpms may not be that much but the hotter you get the combustion chamber and piston and exhaust valve at lower rpms the hotter they will be later when you are approaching peak hp, which will lower your detonation resistance and probably lower the peak safe hp. I think it will be a balancing act - add heat for spooling until you start to see it affect the peak hp. I'm no guru but I have spent a good bit of time making chips for the stock ecm in the 86-87 Buick turbo V6's, and I spent a lot of time tuning the spoolup fueling. I would start logging at idle, mat the pedal, and lift when the boost was over 6-7 psi. Then go back to the log and measure how much time elapsed between the TPS hit and an arbitrary MAF value that was in the 3-4 psi range, and use this time as my spoolup score for that chip. Then look over which timing and fuel enrichment cells were hit, look at the narrow band (and later wideband) O2 values, make some changes, burn a new chip, and do it all over and over again. I also did this starting by brake torquing up against the torque converter to a fixed rpm and boost, and then matting it. Luckily I found that both techniques gave basically the same fuel curve, and that wound up just a bit leaner than the best WOT values at higher rpms. Maybe 12 to 12.5 for spoolup and 11 to 11.5 at WOT above 4000 rpm at 22-23 psi in 3rd gear with 93 octane pump gas. Going leaner did not help spoolup, and if it got lean enough to start detonating during spoolup that detonation would continue into the higher boost and rpm ranges even though the fuel map up there was perfectly fine if it did not start detonating at lower rpms. That's why I say you need to worry about too much heat in the chamber too soon, that will limit you long before the cooling system will.
Interesting. Those are the engines I tend to play with, too, off and on. (More off than on :( )

I think the last one I played with, 23psi was around 1000hp at the crank. The ony before that, 17psi was around 750-800. Large turbos, stroker cranks, GN1 heads.

Spool on the 1000hp engine was ridiculous, but it was also a 10:1 engine bred for race fuel. The turbo response was so rapid that the (owner-installed) blowoff valve would open and chitter at part throttle. Of course, on THAT car, "part throttle" was not much of a concern since it was theoretically able to be driven on the street but for practical reasons it went from trailer to paddock to trailer to garage.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 10:20 pm
by ijames
ptuomov wrote: Mon May 14, 2018 2:53 pm
interesting method and data on optimal AFRs. It seems like you just ended up with the AFRs that would give about the best torque in a normally aspirated engine.

I agree that if you let it start detonating, you'll have to cut boost and ignition advance more to end detonation that started in previous cycles than what would be required for detonation to start without previously detonating cycles. However, if it is not detonating, my logic tells me that heat management at 2500 rpm when the engine is making under 200 hp is a lot easier than heat management at 6000 rpm when it's making four times that. This even considering that the mass flow thru the engine is a lot lower.

Did you find that the car accelerated the fastest below boost threshold rpm with the ignition advance near the knock limit or with the ignition significantly retarded from that? Later ignition timing helping the turbine to some extent.
I spent a lot more time tuning fuel than timing. I tried retarding the timing but since some of those timing cells were also used at light throttle driving at slow speeds it started feeling sluggish, and I never saw any spoolup benefit so I pretty much stayed with the stock timing map until the boost was up, with some adjustment for the torque converter stall speed. A well known Buick chipmaker had told me that he pretty much treated the spoolup region like a normally aspirated engine, then blended into lower timing and more fuel as the boost came up. Since my tuning was almost always aimed at 93 octane I just left the stock table alone until the boost was up past maybe 12-15 psi (the stock maf pegs at 255 gms/sec, about 275 hp, maybe 15 psi boost, and you are at the edge of the timing table by then so tuning is limited). I looked at some commercial race chips for 108 or 116 octane and they were adding in 5-10 degrees of timing below the converter stall and those chips did make the engine hit the converter stall quicker but I couldn't do that with 93 octane. There might well have been gains to be had with timing changes during spoolup but I wasn't able to find any.

Re: Lean spool for turbo freaks

Posted: Mon May 14, 2018 10:25 pm
by ijames
peejay wrote: Mon May 14, 2018 10:00 pm
Interesting. Those are the engines I tend to play with, too, off and on. (More off than on :( )

I think the last one I played with, 23psi was around 1000hp at the crank. The ony before that, 17psi was around 750-800. Large turbos, stroker cranks, GN1 heads.

Spool on the 1000hp engine was ridiculous, but it was also a 10:1 engine bred for race fuel. The turbo response was so rapid that the (owner-installed) blowoff valve would open and chitter at part throttle. Of course, on THAT car, "part throttle" was not much of a concern since it was theoretically able to be driven on the street but for practical reasons it went from trailer to paddock to trailer to garage.
My 23 psi was with a 67 mm p trim turbo and a bone stock long block so a lot less flow than your 23 psi. Front mount, ATR headers which probably hurt more than helped, 72 lb/hr injectors, and a lot of tuning, lots of 12.05 to 12.15's and 3 high 11's on 93 octane and no alcohol injection at 4100 lbs going down the track.

Re: Lean spool for turbo freaks

Posted: Tue May 15, 2018 8:02 am
by ptuomov
Ok it seems that the practical experience is that one should start at the normally aspirated best torque AFR and timing before turbos turn on and then during the spool up period gradually richer the mixture. However it also seems that the engines aren’t very sensitive to the exact AFR in the spool up rpm range. I’ll test that for sure on a hub dyno which can hold the rpm, but that’s some weeks away by my guess.

Re: Lean spool for turbo freaks

Posted: Tue May 15, 2018 9:08 am
by Big Al
Some excess fuel is good in the spoolup area, but the main thing that helps is the ignition timing.
Here in Sweden we have invented the "Turbo start" technique to increase spool "without loosing HP".
It is pretty straight forward, on a specific RPM and boost in the spoolup area of the ignition map where the turbo have some boost but don't have enough energy to really start to spool. There you make the ignition as low as possible, like 0 ignition.
On the road or on the dyno you then feel the engine "missfire" and the spool comes straight away. So you "loose" some HP in that single place but gain spool (and HP ofc) after.
The funny part is that after you have done this right the peak HP increases too!

Re: Lean spool for turbo freaks

Posted: Tue May 15, 2018 10:34 am
by n2xlr8n
ptuomov wrote: Tue May 15, 2018 8:02 am Ok it seems that the practical experience is that one should start at the normally aspirated best torque AFR and timing before turbos turn on and then during the spool up period gradually richer the mixture. However it also seems that the engines aren’t very sensitive to the exact AFR in the spool up rpm range. I’ll test that for sure on a hub dyno which can hold the rpm, but that’s some weeks away by my guess.
Sort of, imo.

Different beast and EMS, but I can send you some logs of my version 8 EJ207 (VF37 @ 27 psi), on E85 if you want. Bone stock it made ~415 ATW.

Same scenario with 93- when it makes >4 psi, I really threw the fuel at it- usually around 11.5:1 and 15-16* of advance.

Re: Lean spool for turbo freaks

Posted: Wed May 16, 2018 9:25 am
by vwchuck
To build boost out of the powerband you add fuel and retard the timing. Then the burn happens in the exhaust system and increases your back pressure to spin the turbo faster almost instantly. This is also know as the anti-lag system in most aftermarket EMS's..

Re: Lean spool for turbo freaks

Posted: Wed May 16, 2018 9:35 am
by ptuomov
vwchuck wrote: Wed May 16, 2018 9:25 am To build boost out of the powerband you add fuel and retard the timing. Then the burn happens in the exhaust system and increases your back pressure to spin the turbo faster almost instantly. This is also know as the anti-lag system in most aftermarket EMS's..
Anti-lag really only works in operating modes where the driver doesn't request maximum torque. That is, during the shift or during part throttle. That's when there's an excess of air flow that can be burned with fuel to run the turbine.

When the driver is requesting maximum torque, there's no such obvious source of excess air. For every oxygen 2 molecule that is burned in the exhaust manifold, there's one less oxygen 2 molecule burned in the combustion chamber and thus less torque applied to the crankshaft. That's the issue I'm struggling with.

Since fuel can't burn without oxygen, just adding fuel to an already rich misture isn't going to cause more reaction. It can slow down the burn, but retarding ignition timing may be a more efficient way to slow down the burn? Or may not.

One logical reason why excess fuel might work is that it may efficiently use heat of the combustion to vaporize and expand, converting heat into pressure, thereby cause the turbine run like a steam engine. But I haven't worked out the physics of that.