There IS such thing as too much meth - or so it looks like..

[themyst]

What tubing are you using? Is it coolingmist provided one?

No I am using labonte tubing which should be the same as DO tubing.

[dzenno@PTF]

Do you actually have problems with timing post shift when you lift off throttle? Or only on wotbox?

i never had timing issues up until the track day and given there was a considerable leak at the solenoid fittings, and there still is a bit of a leak, until I can sort this out I can't say anything for sure other than when there's no leak there were no timing issues wotbox or not

[themyst]

Well, Rob isn't giving up, which I can appreciate. I asked him if it is possible to load an older DME ROM using the AP, he said in theory its possible but requires a rather large firmware revision, would lose the added benefits and programming of the latest DME software and would rather find out why the car is flatlining timing. The latest parameter he wants me to log is "Torque Limit Active". I don't get it to trigger past 0 on stage2+FMIC pump, but it's raining hard here in NYC so I can't really test the race map. Would rather do this after I rule out the meth nozzle being too big first.

I gave him some possible theories, including fuel trims pre and post shift and TCU intervention, but I don't think he's buying any of that.

Either way, I can appreciate the fact he's legitimately interested in the problem instead of simply brushing it off as an inferior meth kit *cough*.

[enrita]

i have never seen my timings flatline on all sorts of configuration. My ECU software is from october/november 2010. My car is a 03-2007.

Well, Rob isn't giving up, which I can appreciate. I asked him if it is possible to load an older DME ROM using the AP, he said in theory its possible but requires a rather large firmware revision, would lose the added benefits and programming of the latest DME software and would rather find out why the car is flatlining timing. The latest parameter he wants me to log is "Torque Limit Active". I don't get it to trigger past 0 on stage2+FMIC pump, but it's raining hard here in NYC so I can't really test the race map. Would rather do this after I rule out the meth nozzle being too big first.

I gave him some possible theories, including fuel trims pre and post shift and TCU intervention, but I don't think he's buying any of that.

Either way, I can appreciate the fact he's legitimately interested in the problem instead of simply brushing it off as an inferior meth kit *cough*.

[themyst]

Mine is from 3/11 which is the same time they revised it for the hpfp campaign.

[turbosetch]

Couple of thoughts her from long time meth user and pro tuner...lol

There is such a thing as to much meth. This persons situation is different than most because he is running 100% meth and not a 50/50 mix or something near that. He is sacrificing the positive effects of the water and just using the octane boost of the meth and some of the cooling effects it has.

That being said when we tune with meth we find that to really benefit from it you have to lean the system a bit and bump the timing...pretty basic stuff. Now if you don't lean the system when you run meth your effectively richening the AFR. This is why people see negative LTFT over time because the systems target AFR is getting screwed up by the additional fuel being added (meth).

What "can" happen when you run to rich or run to much meth is several things...You can get spark blow out or spark quench...You can get poor flame fronts slow flame fronts and generally poor burns in the cyl...You can absolutely create a "knock like" condition from being to rich. I have tuned several cars that started to knock a bit running rich, typical tuning methods tend to say when you knock you add a little fuel, pull timing etc...In some cases if you are rich enough the answer may be to lean the system out a little bit. Now keep in mind you have to know when this is appropriate and when it could lead to bad things..lol..

There are tons of studys and research out there about water/meth systems. They have been in use since the world wars. The general rule of thumb for max power is 5%-10% meth/water to fuel ratio. Meaning for every gallon of gasolene burned you would burn a 10th of a gallon of meth/water. Many WRC cars have run a mix of 100% water even. When you start running 100% meth things change, power can be had this way but the beneficial effects of the water are lost. When running 100% you need to run far higher percentages of it in order to see the same or simular benfits of simply adding a little water. As mentioned in the thread your never going to get the kind of power you would from running race fuel, but meth/water is a good compromise that allows for tuning like your running high octane fuel. Overall the gains are worth the 3 bucks a gallon for meth/water systems.

[dzenno@PTF]

Awesome post man! This is the kind of stuff that gives me a reason to stick around forums honestly and really wish there was more such info shared over time...

I've actually been trying to figure out the mass of fuel injected and logging it somehow..we have fuel trims but fuel mass injected would be a lot more awesome especially when paired with a fuel injector DC or fuel pressure + IPW combo and then hook it into a meth system that references it and appropriately injects (e.g. Aquamist hfs-3)...I really think that today's meth setups have gone from just IAT suppressants at 50/50 mixes and small nozzles to fuel enrichment systems with 100% meth and large nozzles..IMO something has been lost in that translation, in my view that'd be refinement

With a community that's more accustomed to a race towards peak numbers than refinement its really hard to even start discussing such approaches as most users just want easy cheap peak power and vendors/tuners do what most customers want...just nature of things

[themyst]

Couple of thoughts her from long time meth user and pro tuner...lol

There is such a thing as to much meth. This persons situation is different than most because he is running 100% meth and not a 50/50 mix or something near that. He is sacrificing the positive effects of the water and just using the octane boost of the meth and some of the cooling effects it has.

That being said when we tune with meth we find that to really benefit from it you have to lean the system a bit and bump the timing...pretty basic stuff. Now if you don't lean the system when you run meth your effectively richening the AFR. This is why people see negative LTFT over time because the systems target AFR is getting screwed up by the additional fuel being added (meth).

What "can" happen when you run to rich or run to much meth is several things...You can get spark blow out or spark quench...You can get poor flame fronts slow flame fronts and generally poor burns in the cyl...You can absolutely create a "knock like" condition from being to rich. I have tuned several cars that started to knock a bit running rich, typical tuning methods tend to say when you knock you add a little fuel, pull timing etc...In some cases if you are rich enough the answer may be to lean the system out a little bit. Now keep in mind you have to know when this is appropriate and when it could lead to bad things..lol..

There are tons of studys and research out there about water/meth systems. They have been in use since the world wars. The general rule of thumb for max power is 5%-10% meth/water to fuel ratio. Meaning for every gallon of gasolene burned you would burn a 10th of a gallon of meth/water. Many WRC cars have run a mix of 100% water even. When you start running 100% meth things change, power can be had this way but the beneficial effects of the water are lost. When running 100% you need to run far higher percentages of it in order to see the same or simular benfits of simply adding a little water. As mentioned in the thread your never going to get the kind of power you would from running race fuel, but meth/water is a good compromise that allows for tuning like your running high octane fuel. Overall the gains are worth the 3 bucks a gallon for meth/water systems.

What are your thoughts on the impact of water on meth injection? I would venture to say with a large enough nozzle with a sizable water concentration could create a "knock like" condition during the shift event, perhaps?

What would you think of a 80/20 75/25 meth concentration on a nozzle that flows about 900ml/min? Would it impact spark during the shift event on the N54 with an automatic? Not sure if you've worked with the autos on this platform, but during the shift event, timing goes significantly negative, think -18 to -20, and with full meth flow during this shift event, with an 80/20 or 75/25 mixture, do you believe this could cause what you defined as "spark blowout" causing negative timing post-shift?

I uncovered that the ECU perceives my EGTs as too hot for the cats, when it triggers Fuel Mode 6. Is there anything in the above sequence that may cause the EGTs to rise with methanol injection?

You've been insightful, thanks in advance!

edit- Sorry, forgot to mention for stock N54 turbos

[turbosetch]

I should clarify that I have been tuning meth/water systems for 10+ years but my N54 experience has been limited to 3 cars so far. All of which have been JB4 systems.

I don't see why the meth would be spraying during a shift with it properly setup. A truly progressive system would be tuned or setup to lower or stop the pump duty cycles during that period. With a solenoid in line before the nozzle you can stop and start things really fast.

I have run into an issue with our shop car since adding a larger nozzle and temps dropping that causes the car to shut down cyls right at the shift point every once in a while. Although plugs fixed the issue it was not an issue unless meth was flowing...That being said I believe its possible the flow may not be reduced enough before or during a shift causing spark quench and then misfires. Once a cyl goes cold from spark quench it can take a bit to get it cleared out and firing again. The point of misfires must have been right at the threshold and with fresh plugs it helped just enough. Could have just been bad plugs to...<<more research required.

It does seem like with the BMS meth controller that the meth is applied full duty and boost follows. With other systems they use the MAF sensor, boost signal, or fuel injector pulse width to monitor and curve the water/meth flow. I can't really speak for exactly how the BMS tunes are handling the meth but I'm sure Terry could fill you in. If your meth/water system is staying close to linear with your engine load, or fuel consumption, or air flow these are all good ways to stay in tune and balance with the meth/water systems.

So to further on your question I feel its not a timing pull that could cause the spark quench but rather the air flow and load dropping during the shift (I'm pretty sure the throttle closes a bit but the meth keeps on going full swing on logs). Basically the cyl(s) may be getting flooded for a moment. I really don't know about this for sure more testing would be required but it seems to not be a real issue since BMS doesn't address it if it is the case.

On the shop N54 we use a 200+psi pump and a 650ml/min nozzle with the BMS FSB and snowperformance safeinjection box.. The washer tank is filled with -40 Turbo power washer fluid or Mototec -40 concentrate. We had contacted the manufactures of these washer fluid brands to get the ingredients. Basically 40% meth 60% distilled water and blue dye. We have been tuning cars on this mix for several years without issues at all.

For the amount of meth you mentioned 75%-80% I would imagine you would need a larger nozzle like the 900 or a dual nozzle setup to maintain good fine spray at the nozzles.

One trick we use for gauging meth flow is we look for a .5 decrease on the wideband. In other words if the car is showing 11:1 AFR without meth / water we look for 10.5:1 when the system is turned on. From that point we start to adjust timing and lean the system out. Another method is to add water/meth until spark quench is hit then you reduce it a bit below that point. The second method is crude but it works.

On the N54 setups you kind of have an issue with AFR targets because the system seems to be in a constant closed loop mode. No matter what you add or take away fuel wise with meth/water the car will self trim back to the target fuel it has been programmed to target. In this case to tune the meth you would have to pay attention to STFT or LTFT (actually both I guess). This is just a thought and I have never tried it but you would want to note your cars trims with out meth then note the trims after you add meth. My guess is you would want to see the car removing 5%-10% of the fuel in the high load regions to stay on target with meth on. Im kind of spewing numbers off the cuff here so I could be off by a bit but you get the idea.

So what I'm saying here is that there is a way to gauge to much or to little meth/water.

EGTs rising on meth/water setups is backwards, the total effect should be an EGT drop. This tells me the EGT rise is due to some other factors. A quick fact check...EGT will rise on a leaner mix to a point and lower on a richer mix to a point. EGT may rise if you are getting a poor flame front from bad plugs or far too much fuel. EGT will drop with more timing and rise with less timing. EGT will rise with more boost on stock turbos due to back pressures. So a lot of factors are at play here. More logging would be required to figure out why you see an EGT rise. Depending on boost I would suspect its the rise in boost that may be the key factor considering we are spinning these turbos much faster then what they were designed for.

What are your thoughts on the impact of water on meth injection? I would venture to say with a large enough nozzle with a sizable water concentration could create a "knock like" condition during the shift event, perhaps?

What would you think of a 80/20 75/25 meth concentration on a nozzle that flows about 900ml/min? Would it impact spark during the shift event on the N54 with an automatic? Not sure if you've worked with the autos on this platform, but during the shift event, timing goes significantly negative, think -18 to -20, and with full meth flow during this shift event, with an 80/20 or 75/25 mixture, do you believe this could cause what you defined as "spark blowout" causing negative timing post-shift?

I uncovered that the ECU perceives my EGTs as too hot for the cats, when it triggers Fuel Mode 6. Is there anything in the above sequence that may cause the EGTs to rise with methanol injection?

You've been insightful, thanks in advance!

edit- Sorry, forgot to mention for stock N54 turbos

[themyst]

I'm guessing you haven't been working with automatic n54s then? No matter what the meth kit, if you're going WOT throughout the pull you're holding pretty much full boost which means full meth flow.

This is verified even in the jb4/FSB setup.

Interesting point on the egts though. I wish we could log what the ecu perceives as it's egt calculation

[turbosetch]

I agree 100%. There is a big difference between 100% meth and a mix like 50/50 water to meth. Injector duty cycles or MAF is a great way to have a truly progressive system and can really be dialed in.

Our best fuel here in Alaska is 90 octane. The cars we tune that have upgrades like bigger turbos know that meth/water setups are a must to really appreciate the power. BTW I have tuned a car on vp109 race fuel with meth/water turned off. I have taken the exact same car with the same tune and run 90 octane pump gas with a 50/50 mix of meth. The only varibles I needed to adjust in the tune were the AFR targets. Timing and boost all remained the same and no knock events were logged. When I initially tuned with the vp109 I pushed the timing and boost to the knock threshold then backed it up a bit so I know it was an agressive tune. Water/meth can give an easy 20% gain in effective octane "effect" easy. You still will never see the same power as race gas due to the energy simply not being there however its a nice compromise.

Awesome post man! This is the kind of stuff that gives me a reason to stick around forums honestly and really wish there was more such info shared over time...

I've actually been trying to figure out the mass of fuel injected and logging it somehow..we have fuel trims but fuel mass injected would be a lot more awesome especially when paired with a fuel injector DC or fuel pressure + IPW combo and then hook it into a meth system that references it and appropriately injects (e.g. Aquamist hfs-3)...I really think that today's meth setups have gone from just IAT suppressants at 50/50 mixes and small nozzles to fuel enrichment systems with 100% meth and large nozzles..IMO something has been lost in that translation, in my view that'd be refinement

With a community that's more accustomed to a race towards peak numbers than refinement its really hard to even start discussing such approaches as most users just want easy cheap peak power and vendors/tuners do what most customers want...just nature of things

[turbosetch]

The shop car is a 6AT...On the logs if I recall the throttle body is closed a bit by the ecu during a shift regardless of pedal position. Is this wrong?

EDIT post: I went back and looked your right my mistake. New theory time...lol

I suppose the big dip in timing during a shift could account for some issue if the meth is still going full blast.

I'm guessing you haven't been working with automatic n54s then? No matter what the meth kit, if you're going WOT throughout the pull you're holding pretty much full boost which means full meth flow.

This is verified even in the jb4/FSB setup.

Interesting point on the egts though. I wish we could log what the ecu perceives as it's egt calculation

[dzenno@PTF]

Funny you mention Turbo Power -40C washer fluid...it's actually 45-50% meth and 50-55% water and less than 1% other (detergents, dye)..I've run it over 2.5 years on my car and had great results with it, even recommended it to ppl on e90post a LONG time ago as a safe street alternative...I recently started using 100% meth initially with large dual nozzles and now back down to a single CM10 that flows around 650-700ml/min...I'm thinking of going back down to 50/50 for some testing...logs + vbox times really tell the whole story..

[JoshBoody]

You still will never see the same power as race gas due to the energy simply not being there however its a nice compromise.

I'm curious about your comment on energy when comparing meth and race gas. I understand race gas has more energy per volume then meth (and actually less then lower octane gasoline). But this shouldn't make a difference in energy if you are pumping the same air mass... you just add more fuel. Same air mass and same AFR should equal same energy. The reduction in power could have to do with the water content. Anyway curious.

Great reading your posts!!

[dzenno@PTF]

My take on race gas providing more energy is due to many of them being heavily oxygenated, more oxygen more power than with just plain pump gas

[JoshBoody]

My take on race gas providing more energy is due to many of them being heavily oxygenated, more oxygen more power than with just plain pump gas

oh, ok... didn't know that. sorry all for the rant then.

[themyst]

Welp. I logged Knock Voltage on the two noisiest cylinders (1 and 5) and also "Timing Correction" on those two cylinders as a test, and look what I found lol. Right at the point post-shift, the ECU pulls 4-9 degrees of timing causing the impacted gears to flatline. Does this mean the ECU is perceiving the gear change as knock? It appears that fuel trims or whatever else don't mean squat. Does this mean we need to tone down the knock sensor a touch during a gear shift? I wonder if the knock threshold was changed with the later DME revisions?

So many questions left unanswered. On the bright side, at least we can log this kind of stuff with the AP!

[dzenno@PTF]

Great idea about the tranny and knock voltages

[themyst]

Great idea about the tranny and knock voltages

It appears that the knock voltage correlates to a certain amount of noise at a specific RPM, and if the noise exceeds the "safe" parameter it drops timing 3 degrees and adds it back in based on the noise. If the voltage climbs exceedingly fast past the set point, it will drop again. There is a third failsafe in all of this, if the DME detects excessive knock retard it will throw a Glow Ignition code, which is extremely bad. I would venture to guess that these set points are extremely conservative to proactively catch knock retard before it actually happens however. I'm starting to buy into the fact that the DME's knock suppression system is better than I thought. Regardless, I still want a properly timed motor.

Someone who has experience on this topic, PLEASE chime in.

[turbosetch]

Can you note the total timing at shift? Many cars I have tuned will spike the timing on a shift causing actual shift knock. This has been due to the load dropping and the car switching to a higher timing value overall for a split second. The fix has been to go and pull timing out of the high rpm low load cells. Keep in mind the timing pull may be planned to reduce the load on the trans during shift.

I'm sure the knock detection and suppression logic on the BMW is quite sophisticated and impressive. My experience with other manufacturers ECUs shows some pretty nice logic dating back to 2000. I'm going to dig around for some info on this system out of curiosity.

No idea what the N54 timing tables look like yet but if Cobb offers the tuning software soon we will be buying it up for the shop right away.

[Terry@BMS]

Welp. I logged Knock Voltage on the two noisiest cylinders (1 and 5) and also "Timing Correction" on those two cylinders as a test, and look what I found lol. Right at the point post-shift, the ECU pulls 4-9 degrees of timing causing the impacted gears to flatline. Does this mean the ECU is perceiving the gear change as knock? It appears that fuel trims or whatever else don't mean squat. Does this mean we need to tone down the knock sensor a touch during a gear shift? I wonder if the knock threshold was changed with the later DME revisions?

So many questions left unanswered. On the bright side, at least we can log this kind of stuff with the AP!

If you add race gas does it go away?

[themyst]

If you add race gas does it go away?

Nope. I have logs from Atco running MS103 + meth that flatlined as badly. Do you want me to send you chart images that show the differences between knock volt on shifts that flatline versus shifts that don't?

I haven't flatlined in my last 10 recorded logs. Hopefully it's fixed, but it's like a bad pimple, it can come back at any time, which is the frustrating part. I haven't done anything different to warrant any change.

All I did was reset 6AT tranny adaptations using the accelerator method, unmarry and reload Cobb and up the meth mix to 90/10. None of the above should impact the timing flatline.

[themyst]

Can you note the total timing at shift? Many cars I have tuned will spike the timing on a shift causing actual shift knock. This has been due to the load dropping and the car switching to a higher timing value overall for a split second. The fix has been to go and pull timing out of the high rpm low load cells. Keep in mind the timing pull may be planned to reduce the load on the trans during shift.

I'm sure the knock detection and suppression logic on the BMW is quite sophisticated and impressive. My experience with other manufacturers ECUs shows some pretty nice logic dating back to 2000. I'm going to dig around for some info on this system out of curiosity.

No idea what the N54 timing tables look like yet but if Cobb offers the tuning software soon we will be buying it up for the shop right away.

timing on the 6MTs will spike positive upwards of 20+ degrees at times during the shift event, but in the automatics, there is no timing spike. Consistent across 100's of Cobb logs, timing will drop negative and attempt to hit the target after the shift event is complete. Based on these recent logs I've posted, the "lazy cylinders" which don't hit the targeted voltage have their timing "flatline". Why these cylinders are lazy, I don't know. That will take some more digging on Cobb's end.