N54 HPFP VCV Tables

[RSL]

Your spool wasn't nearly as bad in the first log posted and base wasn't jacked up. Whatever you did in between (WGDC Position to WGDC), put it back and then work on WGDC directly if needed. If you're on IJE0S, might consider using RPM x setpoint.

Your HPFP PWM is apparently capped for a bit at WOT. There's a dip right before mass request takes off, not sure if it's related. It requests a lot more injection to offset low pressure/crash operation, just like it reduces injection when pressure is higher than expected. When your idle ramps to 2k psi out of nowhere, it goes leaner and trims kick up to add injection back.

Your idle is high, it's a factor. Your idle PWM is literally as high or higher than mine when driving normally. At normal idle, stock HPFP, mine flips 12 or 30% (switching between pressure and flow) to maintain 670-740psi. Yours is flat at 39% and then moves up slightly when a large mass request comes up and pressure stays there. Clearly, moving back to 38% from 42% isn't enough to let it drop back. Tapping the throttle and letting off closes the VCV a lot more (down to 9%) and the excess pressure disappears. Put your idle back to normal and try it. If you need 950rpm idle, you're probably going to have to tune it to compensate.

 

[MR-KRAKA]

Hi I don’t mean to hijack this thread but I did A LOT of testing with this last year. Credit to Azndevil for posting that xdf earlier in this thread. That’s what got me messing around with this. I was looking to improve flow from the factory setup and I found that there was indeed gains to be made without the Helix or POD. Nothing groundbreaking, the hardware still has its limitations but it was enough to keep me from spending the money on an overdrive. I should have documented more but I will post here where I ended up on a stock HPFP being fed by dual walbro 450’s on a Hobbs switch with E40/Hybrids/25psi. I do have a 500cc water/meth nozzle in the charge pipe for iat’s so that is a factor but it had little effect on the rail pressure when I added it after testing. Just more data for the public eye. I suspect it could also be useful for weak/aging pumps before buying a new one as long as it’s not a full blown hardware failure… Checkbox/slider in MHD WG rattle style? I also believe there is more to be had here if we got it right but after hundreds of pulls I was finally ok with where I was. Here is one of my last logs from last year at the track when 5th gear started telling me no…

Hide legend Reset zoom

I think you guys are missing out on C_FUP_DIF_CUR_VCV_AD (max fuel pressure deviation to allow adaptation) I remember that really helped level things out by taking it up to 1500psi. I also see now that I messed with the other table for homo fuel pressure set point. I can’t remember the effects (if any) but I did leave the changes I made so I will include that table as well.

I saw earlier in here someone getting fuel pump errors after flashing changes, I got that often but you just need to shut the car back off clear the error and restart it’s usually fine after that. Small changes can make a big difference especially with the break points. Again this is what worked for me use these at your own risk! Hope this helps someone.

 

[mj6234]

Hi I don’t mean to hijack this thread but I did A LOT of testing with this last year. Credit to Azndevil for posting that xdf earlier in this thread. That’s what got me messing around with this. I was looking to improve flow from the factory setup and I found that there was indeed gains to be made without the Helix or POD. Nothing groundbreaking, the hardware still has its limitations but it was enough to keep me from spending the money on an overdrive. I should have documented more but I will post here where I ended up on a stock HPFP being fed by dual walbro 450’s on a Hobbs switch with E40/Hybrids/25psi. I do have a 500cc water/meth nozzle in the charge pipe for iat’s so that is a factor but it had little effect on the rail pressure when I added it after testing. Just more data for the public eye. I suspect it could also be useful for weak/aging pumps before buying a new one as long as it’s not a full blown hardware failure… Checkbox/slider in MHD WG rattle style? I also believe there is more to be had here if we got it right but after hundreds of pulls I was finally ok with where I was. Here is one of my last logs from last year at the track when 5th gear started telling me no…

Hide legend Reset zoom

I think you guys are missing out on C_FUP_DIF_CUR_VCV_AD (max fuel pressure deviation to allow adaptation) I remember that really helped level things out by taking it up to 1500psi. I also see now that I messed with the other table for homo fuel pressure set point. I can’t remember the effects (if any) but I did leave the changes I made so I will include that table as well.

I saw earlier in here someone getting fuel pump errors after flashing changes, I got that often but you just need to shut the car back off clear the error and restart it’s usually fine after that. Small changes can make a big difference especially with the break points. Again this is what worked for me use these at your own risk! Hope this helps someone.

No hijack at all. If you have the address for C_FUP_DIF_CUR_VCV_AD - particularly for IJEOS, I would appreciate it. Interesting on the increase in DC beyond 77% in the stock table. I can do some testing on that as well. Based on your experience, I feel like I am likely limited on the LPFP side at this point. That log looks great, especially for stock HPFP!!!

 

[MR-KRAKA]

No hijack at all. If you have the address for C_FUP_DIF_CUR_VCV_AD - particularly for IJEOS, I would appreciate it. Interesting on the increase in DC beyond 77% in the stock table. I can do some testing on that as well. Based on your experience, I feel like I am likely limited on the LPFP side at this point. That log looks great, especially for stock HPFP!!!

View attachment 67528

 

[mj6234]

Your spool wasn't nearly as bad in the first log posted and base wasn't jacked up. Whatever you did in between (WGDC Position to WGDC), put it back and then work on WGDC directly if needed. If you're on IJE0S, might consider using RPM x setpoint.

Your HPFP PWM is apparently capped for a bit at WOT. There's a dip right before mass request takes off, not sure if it's related. It requests a lot more injection to offset low pressure/crash operation, just like it reduces injection when pressure is higher than expected. When your idle ramps to 2k psi out of nowhere, it goes leaner and trims kick up to add injection back.

Your idle is high, it's a factor. Your idle PWM is literally as high or higher than mine when driving normally. At normal idle, stock HPFP, mine flips 12 or 30% (switching between pressure and flow) to maintain 670-740psi. Yours is flat at 39% and then moves up slightly when a large mass request comes up and pressure stays there. Clearly, moving back to 38% from 42% isn't enough to let it drop back. Tapping the throttle and letting off closes the VCV a lot more (down to 9%) and the excess pressure disappears. Put your idle back to normal and try it. If you need 950rpm idle, you're probably going to have to tune it to compensate.

I was trying to get away from the SMFW chatter at idle. I will flip it back and see how things look on that front!

 

[mj6234]

On another side note, I noticed AZDevil's post had a few other things defined regarding startup. I am going to see if I can get rid of the startup chatter that happens for a few seconds based on what he has in there.

I will say, I am truly humbled by the sharing and feedback. It is neat to see how much others have already figured out that is still new to me.

 

[KClemente]

I also see now that I messed with the other table for homo fuel pressure set point. I can’t remember the effects (if any) but I did leave the changes I made so I will include that table as well.

The other homogenous fuel setpoint table (ip_fup_sp_hom) is from the Continental layer and has no effect as a bit (S_PRAIL_SW) is flipped to 1, making the DME use the BMW layer map (KF_PRAIL_H). Though no harm, no foul in changing the Continental layer

I've had C_FUP_DIF_CUR_VCV_AD set to 800 psi for a long time and figured it was a good number to set as my fuel pressures never really deviate more than 800psi. But it definitely deviated more than the stock 580psi

For ethanol blends, I'd probably set it to 1000 psi

Also going pretty much 100% duty cycle is an interesting move and may work for people who are trying to squeeze the most out of their pump and is probably the biggest factor in your pump capacity

 

[mj6234]

Couple small updates.

(1) Bought a 295 pump and replaced the 274 in my bucket. Doing a little driving to make sure I didnt screw up anything then will relog on a new pull. My guess based on last log was out of LPFP, I guess we will see.
(2) After @RSL suggestion, I reset my idle. No pressure racing situation since then. I also noticed the ~33% pump DC at idle cycling down to ~12% pump DC randomly at idle, which I didnt have before. I also noticed that my current request was all the way down to 1.1X, which seems to be even lower than I expected based on the L/H to current tables. Possibly changing (3) also had an impact here?
(3) After setting the max pressure deviation table to 1000psi, I can 100% say the HPFP pressures are way less spiky. Curious to see how much smother things are in a log, but this one seems to greatly affect the behavior at WOT.

 

[RSL]

2 is similar behavior I see stock idle/HPFP. I assume there's either a current or flow threshold/gradient limit that causes the mode switch to bump pressure on it. Real-time, this looks like rail floating between 660-750psi. I'd hazard a guess if pressure control side was adjusted a bit, it wouldn't drop enough to need the bump.

If you're going bump idle back up, you'll probably want to find either what's causing the mass request increase or where to reduce HPFP, so it doesn't stick high once the request drops.

Of course, if 3 did have anything to do with it, you may not need to do anything.

 

[mj6234]

Couple small updates.

(1) Bought a 295 pump and replaced the 274 in my bucket. Doing a little driving to make sure I didnt screw up anything then will relog on a new pull. My guess based on last log was out of LPFP, I guess we will see.
(2) After @RSL suggestion, I reset my idle. No pressure racing situation since then. I also noticed the ~33% pump DC at idle cycling down to ~12% pump DC randomly at idle, which I didnt have before. I also noticed that my current request was all the way down to 1.1X, which seems to be even lower than I expected based on the L/H to current tables. Possibly changing (3) also had an impact here?
(3) After setting the max pressure deviation table to 1000psi, I can 100% say the HPFP pressures are way less spiky. Curious to see how much smother things are in a log, but this one seems to greatly affect the behavior at WOT.

New log. Still gets a little spiky up top. Clearly the 274 was out of fuel and the 295 is not. I should have ordered the bucket with the 295 from the get go, but I guess I have a 274 if I ever need to go stage 3 LPFP.

I know the tune itself still has some issues, and I will keep working on it more. But honestly it drives well, feels reasonably fast, no codes, no throttle closure, no other totally weird behavior. I am not going for any world records, just want to learn and have something reasonably safe/fun to drive. I still need to replace solenoids at some point, and see if it helps spool.

Side note - setting the "Fuel pressure target during ign off (Homo-Cold)" from 2900 to 2200 pretty much eliminated all chatter at startup. I didnt change the current from the stock value of 1.8A.

I set the 'adaption' value identified above to 1000, but may increase based on the other guidance provided.

datazap.me | mj6234 | Log 1652051322

datazap.me

 

[MR-KRAKA]

New log. Still gets a little spiky up top. Clearly the 274 was out of fuel and the 295 is not. I should have ordered the bucket with the 295 from the get go, but I guess I have a 274 if I ever need to go stage 3 LPFP.

I know the tune itself still has some issues, and I will keep working on it more. But honestly it drives well, feels reasonably fast, no codes, no throttle closure, no other totally weird behavior. I am not going for any world records, just want to learn and have something reasonably safe/fun to drive. I still need to replace solenoids at some point, and see if it helps spool.

Side note - setting the "Fuel pressure target during ign off (Homo-Cold)" from 2900 to 2200 pretty much eliminated all chatter at startup. I didnt change the current from the stock value of 1.8A.

I set the 'adaption' value identified above to 1000, but may increase based on the other guidance provided.

datazap.me | mj6234 | Log 1652051322

datazap.me

View attachment 67894

Not sure what your goals are but why not lean on your capacity and increase ethanol content? For the spike take a look into spool mode/timing/vanos make sure it blends into you main tables smoothly. It does look like you still have lean spool enabled so that’s probably your culprit.

 

[mj6234]

Not sure what your goals are but why not lean on your capacity and increase ethanol content? For the spike take a look into spool mode/timing/vanos make sure it blends into you main tables smoothly. It does look like you still have lean spool enabled so that’s probably your culprit.

I will definitely be doing that. Raising the boost as well. I want to see how far I can make it on straight E85, even though I know it won't make the same power there as on an e-blend.

This is the first time I have run 22psi up top, but definitely has more in it now that I changed the pump. Ultimately I thought 575whp was a decent goal without getting into the danger zone and/or get to the point where I'd dump a ton more money for an increase in power.

 

[mikegsx]

As someone who recently installed a helix, this thread has been super helpful. I was able to get the chatter down by lowering cold fuel pressure target to about 2200 psi. it only chatters for the first 10 seconds and that's fine with me. I really want to try and adjust the max fuel pressure deviation for adaptation but run into issues when i try to define the parameter. I would like to smooth out the rail pressure during WOT. whenever I make the vcv flow adjustments though I seem to hit the pressure relief at 18 MPa.

 

[mj6234]

Couple other updates on things I have tried. With the AC on, I have gotten the rail pressure surges I described before. I assume a little more load with the AC on pushed it into a different part of the VCV flow request map. I had previously reduced the amp request in the 1 & 2 L/H cells and got to a point where the car would start, but then die off pretty quickly and throw HPFP codes.

I rethought things a little, and wondered if the VCV needed some minimum current to open at all, and that was near the value BMW put in the stock table (and when I lowered the 1 L/H cell to 1.15 or whatever, it wasnt enough to open the VCV, and that caused the motor to die). Also note the stock value in the 0 L/H cell is 1.2, which I forgot I had previously changed to 1.0 when messing with things. So I redid things on the low end with that in mind, and seem to finally have gotten rid of the racing rail pressure at idle. My next test is to raise the idle when the AC is on and see if it is gone in that case as well (and limit my SMFW chatter).

I think from the 30 L/H cells on, they probably need to be relooked at. I still have odd drop in rail pressure during spool that others have noted with the Helix. I feel like some of the values are still too high, but I didnt have much opportunity to reduce them further without first reducing the values more in the left part of the table.

Last thing I tested was the fuel pressure target on startup. I do still have some minor chatter on startup while the high pressure settles. Previously I reduced the stock value from 2900 to 2200,which helped. I tried 1900, 1800, and 1700 as well. 1700 for some triggered some sort of safe mode where the rad fan went to max, I got a 29E1 code, and it was clear the car was dumping fuel. I recreated it a couple of times, but not sure what caused it. I ended up going back to 1900 and 1.7A for the startup current and seems OK. Probably 3-5 seconds of chatter is all I get now.

I will report back on the racing rail pressure and raised idle stuff in the future. Also if I get to some other good result by continuing to mess with the flow/current request table.

 

[mj6234]

One more update, I did end up having the racing rail pressure at idle occasionally with the above. I made more changes and raised the idle and no issues at all for the last few weeks. I did have to update the PWM DC down a couple of % as well as the current table down another 0.01. As mentioned before, it doesnt seem like you can go below a certain amount in the current table in the low end.

What I noticed when logging at idle was the PWM DC value was below the value set in the current to PWM table. I assumed that meant the PID was trying to pull it down, so I dropped the middle part of the table by 2%, and it is more in line (although still seems to be lower than what is in the table).

At any rate, this may end up being my landing spot. The only issue outstanding is that momentary downward spike around 3000 rpm when the turbo is spooling. It isnt noticeable when driving, although I still suspect the HPFP tables are too high in that area (or leading up to that area) and it is the DME intervening. I would summarize my thoughts on messing with these tables as approaching it in 2 steps: (1) try and fine tune the flow to current table first then (2) make small adjustments to the current to PWM table 2nd.

As often as I had the racing rail pressure, I just cant believe other users havent had it as often as me. Maybe they do, but don't notice? Not sure.

 

[mikegsx]

yeah so i've been running my a/c more lately and I have had a few times where the fuel pressure seems to be running high at idle too but I haven't had a chance to do anything about it yet.

 

[Nosrok]

Does these VCV tables work with MHD plus? Finally got my car back up and running but I'm getting a 29F1 code, wasn't an issue with previous tune but I used a base to rebuild for MHD plus. I see it had some VCV tables but missing a bunch as well.

Setup is double shotgun, Bottom mount single G30-900.

 

[MR-KRAKA]

Does these VCV tables work with MHD plus? Finally got my car back up and running but I'm getting a 29F1 code, wasn't an issue with previous tune but I used a base to rebuild for MHD plus. I see it had some VCV tables but missing a bunch as well.

Setup is double shotgun, Bottom mount single G30-900.

Yes, I believe they included the more commonly used HPFP tables in their XDF’s when they released it. Otherwise if you have an older XDF with a specific table you’re looking for it can be easily added to your copy of the latest XDF. You could also just use that old XDF to make the changes you want to your bin in relation to the HPFP and save it. Afterwards return to the current MHD+ XDF. If it’s located in a part of modified code you will know pretty quickly when you open the table in TP.

 

[BEAT_ITN54]

Hi I don’t mean to hijack this thread but I did A LOT of testing with this last year. Credit to Azndevil for posting that xdf earlier in this thread. That’s what got me messing around with this. I was looking to improve flow from the factory setup and I found that there was indeed gains to be made without the Helix or POD. Nothing groundbreaking, the hardware still has its limitations but it was enough to keep me from spending the money on an overdrive. I should have documented more but I will post here where I ended up on a stock HPFP being fed by dual walbro 450’s on a Hobbs switch with E40/Hybrids/25psi. I do have a 500cc water/meth nozzle in the charge pipe for iat’s so that is a factor but it had little effect on the rail pressure when I added it after testing. Just more data for the public eye. I suspect it could also be useful for weak/aging pumps before buying a new one as long as it’s not a full blown hardware failure… Checkbox/slider in MHD WG rattle style? I also believe there is more to be had here if we got it right but after hundreds of pulls I was finally ok with where I was. Here is one of my last logs from last year at the track when 5th gear started telling me no…

Hide legend Reset zoom

I think you guys are missing out on C_FUP_DIF_CUR_VCV_AD (max fuel pressure deviation to allow adaptation) I remember that really helped level things out by taking it up to 1500psi. I also see now that I messed with the other table for homo fuel pressure set point. I can’t remember the effects (if any) but I did leave the changes I made so I will include that table as well.

I saw earlier in here someone getting fuel pump errors after flashing changes, I got that often but you just need to shut the car back off clear the error and restart it’s usually fine after that. Small changes can make a big difference especially with the break points. Again this is what worked for me use these at your own risk! Hope this helps someone.

do you happen to have the Max fuel pressure deviation defined for ina0s?

 

[MR-KRAKA]

do you happen to have the Max fuel pressure deviation defined for ina0s?

0x46464 with a conversion of (X*5.306721599145)*0.014503773773 stock should be 580psi