Hydra 5700RPM hesitation, cannot tune out

[jblaine]

I have an intermittent hesitation around 5700-5800RPM that I cannot tune out. It happens about 50% of the time.

It "feels like" a misfire.

AFRs are steady at 11:1

Nothing at all shows as far as high knock voltage.

I have tried lowering boost to 13psi.

I have tried lowering timing there to as low as 15, incrementally.

Anyone have a clue?

* Registered users of the site do not see these ads.

[ahains]

Same RPM regardless of gear?
Electrical gremlins? Have good grounds?

[jblaine]

I'll do some testing in a gear other than 4th and note the RPM.

Grounds are better than stock and recently checked. I will recheck though - easy enough.

[Element Tuning]

Lower the Dwell by .6 across the board and smooth out the dwell trim in that RPM range. The older WRX coil packs are less sensitive to dwell but the new coil packs used on the STI are very sensitive and lowering should help.

I can also just update your current map if you email it to me.

Thanks,
Phil
www.elementtuning.com

[jblaine]

Well at least it's comforting to know that, last night while in bed staring at the ceiling going over this in my head for an hour (after doing another 90mins of late night tuning) that my hypothesis about coil packs wasn't too far off!

I'm getting just dangerous enough to pull a rabbit out of my hat now and then.

I will email you the current map within the hour, Phil.

[ride5000]

Originally Posted by Element Tuning

Lower the Dwell by .6 across the board and smooth out the dwell trim in that RPM range. The older WRX coil packs are less sensitive to dwell but the new coil packs used on the STI are very sensitive and lowering should help.

I can also just update your current map if you email it to me.

Thanks,
Phil
www.elementtuning.com

jeff, do you have these new sti coil packs?

phil, lowering dwell reduces conduction angle, correct?

and this will reduce spark energy, correct?

ken

[Element Tuning]

Originally Posted by ride5000

jeff, do you have these new sti coil packs?

phil, lowering dwell reduces conduction angle, correct?

and this will reduce spark energy, correct?

ken

I don't know the answer to your question as it pertains to conduction angle. I do know that the older coil packs could take up to 5 ms total dwell without much trouble but the newer coil pack are happier with 3-4 ms of total dwell time. What’s important to remember is that you only need enough dwell time to fully charge a given coil. Adding more time doesn’t necessarily benefit anything and in fact I usually find it can burn out coil packs and cause misfires. Why the misfires? I’m not 100% positive if this is due to overheating or possible spark jumping. I can tell you with conviction that I have not lost power by reducing the dwell times of the STI as I have tested the lower settings on the dyno. We usually make more power because it stops misfiring.

Thanks,
Phil
http://www.elementtuning.com

[silvialost]

thats a lean cut at 11.1 afr.
my 95 ej20g does that.
subys run best any where in the mid 10s for the afr, at least all the ones i have seen on the dyno.
add some fuel at the rpm where the hesitation occurs. that should smooth it out.

[rjrutzky]

Originally Posted by silvialost

thats a lean cut at 11.1 afr.
my 95 ej20g does that.
subys run best any where in the mid 10s for the afr, at least all the ones i have seen on the dyno.
add some fuel at the rpm where the hesitation occurs. that should smooth it out.

Why didn't I think of the old 11.1 LEAN fuel cut? DUH

[jblaine]

Originally Posted by ride5000

jeff, do you have these new sti coil packs?

phil, lowering dwell reduces conduction angle, correct?

and this will reduce spark energy, correct?

ken

Nope, 2002 WRX coil packs.

[fragment]

Hmm. I'm going to have to try this. I get a misfire and sometimes even poping exhaust at 6800 RPM regardless of gear.

[ride5000]

Originally Posted by Element Tuning

I don't know the answer to your question as it pertains to conduction angle. I do know that the older coil packs could take up to 5 ms total dwell without much trouble but the newer coil pack are happier with 3-4 ms of total dwell time. What’s important to remember is that you only need enough dwell time to fully charge a given coil. Adding more time doesn’t necessarily benefit anything and in fact I usually find it can burn out coil packs and cause misfires. Why the misfires? I’m not 100% positive if this is due to overheating or possible spark jumping. I can tell you with conviction that I have not lost power by reducing the dwell times of the STI as I have tested the lower settings on the dyno. We usually make more power because it stops misfiring.

Thanks,
Phil
http://www.elementtuning.com

i wasn't sure of the units of the hydra's dwell parameter, and thought for a moment that it was in degrees of crank rotation--that's all i meant by conduction angle. from what you've posted it in straight ms.

that's very intriguing about the longer dwell actually precipitating misfires. one would assume that the opposite would be true, but from what you're saying it seems there's a window of dwell that works best. too long or too short and you get issues.

hmm. something is starting to make some sense here. i have a lot of experience with old antiquated vacuum tube amplifiers which use transformers to couple the high impedance tube circuits to the low impedance speaker loads. a coil is a transformer.

it is absolutely true that once the magnetic circuit of the coil is saturated you will not get any more transmission of energy when the current is interrupted. in fact, depending on the core materials of the coil itself (ie, steel, ferrite, etc), you will likely see a DROP in magnetic permeability once a sufficient level of flux is passing through the core and saturation occurs. a lot of core materials exhibit this tendency. there's a sweet spot in terms of flux vs. permeability where very low and very high flux levels show a dropoff.

the permeability is like a measure of the magnetic conductivity of the material. high permeability means high efficiency in terms of the magnetic portion of the coil. this is a "good thing" and is probably a direction in which the electrical engineers at FHI are going. in other words the newer coil packs are probably more efficient and thus saturate at lower flux levels, and the flux is determined by the # of turns on the coil and the amount of current passing through the coil. the dwell directly sets the amount of current.

so there is some logic behind the "sweet spot" for dwell duration and ultimate spark energy. more current doesn't always = more spark energy because there is an "imperfect" magnetic core material involved that has a varying permeability wrt flux density.

hth
ken

(for the curious, i have a good article on electromagnetics on my website written by an EE friend of mine, henry pasternack. http://ken-gilbert.com/techstuff/magnetics.html)

[n2xlr8n]

Originally Posted by ride5000

i wasn't sure of the units of the hydra's dwell parameter, and thought for a moment that it was in degrees of crank rotation--that's all i meant by conduction angle. from what you've posted it in straight ms.

that's very intriguing about the longer dwell actually precipitating misfires. one would assume that the opposite would be true, but from what you're saying it seems there's a window of dwell that works best. too long or too short and you get issues.

hmm. something is starting to make some sense here. i have a lot of experience with old antiquated vacuum tube amplifiers which use transformers to couple the high impedance tube circuits to the low impedance speaker loads. a coil is a transformer.

it is absolutely true that once the magnetic circuit of the coil is saturated you will not get any more transmission of energy when the current is interrupted. in fact, depending on the core materials of the coil itself (ie, steel, ferrite, etc), you will likely see a DROP in magnetic permeability once a sufficient level of flux is passing through the core and saturation occurs. a lot of core materials exhibit this tendency. there's a sweet spot in terms of flux vs. permeability where very low and very high flux levels show a dropoff.

the permeability is like a measure of the magnetic conductivity of the material. high permeability means high efficiency in terms of the magnetic portion of the coil. this is a "good thing" and is probably a direction in which the electrical engineers at FHI are going. in other words the newer coil packs are probably more efficient and thus saturate at lower flux levels, and the flux is determined by the # of turns on the coil and the amount of current passing through the coil. the dwell directly sets the amount of current.

so there is some logic behind the "sweet spot" for dwell duration and ultimate spark energy. more current doesn't always = more spark energy because there is an "imperfect" magnetic core material involved that has a varying permeability wrt flux density.

hth
ken

(for the curious, i have a good article on electromagnetics on my website written by an EE friend of mine, henry pasternack. http://ken-gilbert.com/techstuff/magnetics.html)

I would think heat= lower Q, less energy transfer, more heat in the coil=open circuit. No?

S.

Edit: BTW, that's a great read. I've worked in industrial RF and electromagnetics (MV/MeV) for 15 years and loved that. Your friend is a talented writer.
s

[ride5000]

hey steve,

henry is a smart guy. i ran into him first on rec.audio.tubes around '98 or so. worked for bell research for years. funny how some of the smartest guys i know worked for the phone company... or at least used to.

as far as the effects of heat on the electromagnetics.. you're right that as the heat goes up, the DCR of the windings will too, and that will change the Q... BUT whether or not that makes a difference in the spark energy depends on whether we're talking about a tuned circuit in the first place. i really don't know whether or not they're taking advantage of a sort of multi-spark bounce that could occur if the circuit oscillates.

i've got an LCR meter on my workbench so next time i change plugs i'll get some numbers. hmm, should be in around 8 months now..

[bboy]

I had a similar problem. When I stepped on the gas (WOT) at say 3500 RPM it would misfire right around 5600 RPM. If I stepped on the gas at 2000 RPM, it would NOT misfire. I also noticed when logging that depending on when you jumped on the gas the AFRs were different.

The solution was to lower the throttle pump setting in the 3500, 4000, 4500, etc.... where ever I got misfires when I got on the throttle.

[jblaine]

That doesn't make any sense to me at all since Throttle Pump is a dynamic tip-in enrichment that affects your fuel for about .5 seconds and stops.

Strange.

[bboy]

What if it's more than 0.5 seconds. I agree it does not make sense. I'm reporting what I observed.

Maybe the dynamic enrichment takes into account the throttle pump settings. There is so much we don't know about the algorithms in the Hydra.

I tried 'dwell' as well to no effect. I am curious that we both seem to observe it at the same RPM. For me it was like clockwork.

[jblaine]

Well, I've got Phil's adjustments to my Dwell and Dwell Trim and did some pulls late last night. Too early to tell for me. I'm sure in a few days I will have success or failure to report.

"There is so much we don't know about the algorithms in the Hydra."

Ain't that the truth I'm still trying to make clear and comprehensible definitions of many settings.

[bboy]

OT

Jeff, I have not forgotten about your request for the knock listening via computer set up. I've got it worked out on my Mac and I need to return to making it work on the PC. I found some freeware sound processing software but I still need to test it. Trent

[jblaine]

Ah, you DID get that email Cool.

The EE part is my biggest stumbling block. I am completely electronics clueless.

[x99percent]

Originally Posted by bboy

I had a similar problem. When I stepped on the gas (WOT) at say 3500 RPM it would misfire right around 5600 RPM. If I stepped on the gas at 2000 RPM, it would NOT misfire. I also noticed when logging that depending on when you jumped on the gas the AFRs were different.

The solution was to lower the throttle pump setting in the 3500, 4000, 4500, etc.... where ever I got misfires when I got on the throttle.

How much does the Dynamic Enrichment Coefficient (Settings, Injection) matter here?

[jblaine]

IMO, neither Dynamic Enrichment Coefficient nor Throttle Pump should have anything to do with me going WOT at 3200 and then hesitating when I hit 5700 many seconds later (4th gear).

My understanding is that they are immediately-acting and incredibly brief.

[bboy]

For me it was "rich" misfires from too much fuel enrichment. I wish I had the logs to prove the AFR change, but I don't anymore. It was almost like the throttle pump burst and then decayed. I know it's crazy!!

[bboy]

Oh one more thing. New plugs briefly cured the problem, but then it returned. That could be a clue that goes with the 'dwell' theory, and is likely why I tried changing dwell.

[Element Tuning]

Running leaner, adding timing, and changing spark plugs are all basic tuning strategies to reduce misfires.

Thanks,
Phil
www.elementtuning.com