How to scale the HPFP pressure sensor
 

Okay, so there's no way to just solder in a diode or resistor to drop the reported hpfp pressure sensor voltage, because there is a pull up resistor on the ecu side, and the sensor is effectively "sinking" the current. All a diode will do is stop all current flow, and make the ecu see 5V and lean the hell out of the fueling. And if you reverse the direction of the diode, it'll just do the exact opposite of what your trying to achieve.

Resistors don't work cause all they really do is alter how much current the sensor is sinking.
.
.
.
Long story short, you need to supply a scaled voltage to the ecu, and for this, we use an op-amp.

I noticed that there was more than just a pull up on the ecu side, but also a significant resistance to ground as well, basically a voltage divider. I figured that this could be some sort of fail safe for a failed sensor. It's about 4.6k to the supply and 6.8k to gnd, which would result in a voltage of 2.92V reported to the ecu if the sensor were "disconnected". I figure this was put there for a reason, so i mimiced it in my circuit.

http://i359.photobucket.com/albums/o.../schematic.jpg

There's not much to the circuit, for those DIY types, just a non-inverting op-amp with a adjustable voltage divider on the input.


The circuit limits the amount of adjustment you have. So you can turn the scaling all the way off, basically running normal. Or you can turn it all the way up, but it won't let you turn it into the danger zone, lol. That's what the "limiting resistor" does.


Here's a plot of the voltage output with the pot set to minimum (no scaling):
http://i359.photobucket.com/albums/o...inimumplot.jpg

You can see that Vout clips at just over 4V. This isn't an issue cause the ecu will never command more than 4V.

Here's a plot of the voltage output with the pot set to maximum (full scaling):
http://i359.photobucket.com/albums/o...aximumplot.jpg

The full scaling is roughly 22% increase across the board.

Here's a vid of the circuit working on a bench, with a power supply acting as the sensor input:

This is when i was about to put it on the car:
http://i359.photobucket.com/albums/o...itecircuit.jpg

And it on the car:


Here's a vid with minimum setting and an idle injection pulse width:

Here's a vid with maximum setting and an idle injection pulse width, you can see that the injection duration is indeed brought in. I'll do much more testing when i get my upgraded internals in and can actually drive the car without worry :) .




I mentioned in the other thread, if it proves to be safe on my car, and enough people want to do it, but don't want to build their own, i can order some pcb's and parts and stuff.


*disclaimer*: Don't do this to your car unless your willing to assume all responsibility of damage incurred from this "mod". This is thread is not intended to be an instructional "how-to". But... if you do choose to do it, GL and let us know how it works :)

or you could just use the standback - awesome work tho - you're a really smart dude! - you think someone could use your chart there to help interpret what they would need to scale w/ the sb

Possibly. The charts are just simulations. Easiest to just use the fuel trims and re-cal the maf.

Amazing how the standback had this feature all along, lol. When i get my upgraded fp, i'll do a couple dyno pulls comparing with and without the higher pressure. I'll also be doing alot more injector logging.

awesome work man, much respect for you

put me down for one :D

That looks super legit man. Awesome job. How ironic that you figure out how to juice the fuel pump when yours is down for the count.

I'm looking forward to seeing how far this can take the car. I've got a boost gauge you can have if you want to tune for 26 psi when your ish is back together.

Let me test it out to completion first, and then maybe i'll be willing to order some boards and stuff. Or i may decide to change the circuit up depending on tuning etc. May be nice to use things like boost, rpm, maf etc along with the scaling, kinda like a slimmed down SB. We'll see.

LOL, i have one staring at me right now along with a block8head gauge pod! Let's hope this indeed helps, and doesn't end up just fucking my injectors up and ruining my motor, lol.

I vote we have mark test this out, I think he needs it the most

I think he has the standback, right? Which is actually probably more capable than this circuit, but then again, this circuit is much cheaper than a sb, lol. I agree though, next step is a dyno A-B comparison with increased fuel pressure as only change. I'll def give it a go if nobody else does in the next couple weeks.

you're right, he does have a sb, but if he could also try the circuit and compare/contrast...

mark...you know you want to!

silverdemon already did some real world testing - in the other thread - he got his car to dip down to 8 afr - scary stuff! - it just needs the hand of tuner - even tho i'm bailing on the car i'm still pumped to see what mark can do with way more fuel and the massive air coming form the freggin gt40- can't help but plug - i haz sb for sale cheap - see part out thread

Excellent work Dustin - glad you can get the circuit together in 1 day.

EDIT - couple of notes:

I like that you used a high limiting resistor to not load the output of the sensor.

In terms of mimic-ing the 4.6k - 6.8k divider for some sort of failsafe (the 2.92V) - it's probably not necessary as that will be scaled if the pot is anything but 0 and the ECU won't see that voltage value even if the sensor is disconnected.

+1 on the nice work.

I would like to see how much the injector pulse width changes in the high RPMs.

Only changing the pressure and nothing else will not give the desired results, AFR's are directly related to increasing the pressure (common knowledge). Again, I am not that good of a tuner, just reporting what happened on my test yesterday. In fact the car laid completely down when it targeted the increase pressure, AFR's went way low. I think where I went wrong was I had it scaled to the TPS (did not realize until after I had the logs), where I should have scaled it to the MAF, and made a little more of a linear increase in the pressure as MAF voltage increased, then went after trying to scale the MAF sensor to keep the AFR's in check......please correct me if I am wrong since I am really not sure.

SilverDemon, we're all aware, especially djuosnteisn, that the increased pressure needs to be tuned for, and that, by itself, it may not lead to any increase in power. But, if it works as expected, with no surprises, it's the the answer to increasing fueling needs to make big power on the MZR.

Dustin, I was just thinking of the tour of your shop you gave me a while back. You need to hijack some resources to build some slick as shit, plug and play HPFP scalers. You're perfectly set up to do it. First one to do it makes the big bucks.

Fuel Pressure Tuning - Mazda6 / Atenza

dustin is right and dada is testing it out right now as well

There isnt a vid of this, but supposing that the 22% pressure increase does corrilate to ~11 lower injector PW, then we should get 11% more power.

Our injectors max around 420 crank hp, so 420*1.11=466 crank hp *.8 = 372whp @ 7000rpm.

Or, if we lower the rpm peak to 4500rpm with the formula 4500/7000=.642, 1 -.642 = .358 + 1 = 1.358 * 372whp = 505whp @ 4500rpm.

Or, if we lower the rpm peak to 5500rpm with the formule 5500/7000=.785, 1 -.785 = .215 + 1 = 1.215 * 372whp = 451whp @ 5500rpm.

Its funny that their dyno from last week when they got this fuel pressure idea all on their own, is dated yesterday.

That is rather convienient..

Yeah, but i figured seeing at least a "scaled" version of the divider voltage was better than seeing 0V, lol.

I'm stoked that it seems to work. Wish i could drive around with adam and the scope now, lol, but we will as soon as i get my new pump.


In the mean time, i was thinking of building up like 5 or so, and letting some people beta test if they want, at their own risk! Haha, can't add enough disclaimer to this, cause it's definitely experimental research at this point.

The 5 wouldn't be beautiful PnP units, they'd be ugly like mine, haha, and require soldering probably. And i honestly think they should go to some people pushing the limits of our power thus far, like socks, and definitely mark if he wants to try. And anyone else really.

I may ask for some money for the time to build them and parts and shit.

I'll take one. :)

This is not true. Hence, why people (like me right now) are running a stock FP and dropping to 400psi or less WOT but, still holding AFR's in the low 11's and even 10's. There is a table that we can't see yet with ATR that must control this compensation, it's been talked about some in a couple other threads.

Sorry, I have done a little testing where all I did was increase pressure and the AFR went to 8.xx

A couple of things to keep in mind.

The car will target certain AFRs using more variables than just pressure, MAF, and boost. It's a little more complicated than that.

The AFR you read from the O2 sensor does not tell you the whole story about what is going on in the combustion chamber.

Exactly, but how else are you going to measure what is going on inside the combustion chamber?

I am not following, if this mod does the same thing that the SB can do, how come nobody with a SB has made 450whp yet?

i'll take when ur ready to build

Directly through instrumentation that is too expensive for the average guy to have.

Indirectly you have to get creative and make some inferences :)

Watching EGT's as well as the wideband is the best method any of us are ever going to have... Unless you want to mortgage your house and buy some real testing equipment..

Yea it seems that the ECU probably targets AFR first(or close to first) then makes changes to other variables to hit that target. So less pressure= less fuel in per stroke= less air in/ less boost to compensate and hit the target AFR. Also goes the other way, which is why people can make more power with an upgraded pump at 1800+ psi at whatever target AFR they want.

The standback has apparently had the capability all along, but hasn't been expirimented with up until now.

To be honest I'm way more interested in what dustin can come up with than p3.

Well I have an EGT gauge, and I read the factory wide band, and I could always go back to old school and read the plugs......

because wiring this circuit into your fuel system is a much easier/safer solution then opening a laptop?

i think the answer is that jordan isn't with cpe anymore - from my understanding lou was the one that origionally cracked the ecu for the ms3 and developed the sb - that's not to say that he was the one that got into it's full capabilities - i think jordan had a hand in tuning that robbie guy's car in college park md w/the sb - and to date it believe it's still the record for most hp - like 410 w/out nos or meth - i would have like to have seen the tables on that tune -
i think the real question is who has gone to cpe/p3 and really ponied up all money they want for all the stuff they can do - and what can they do now that jordan is gone and just the folks that are using what he developed are left?

so is that all that's left at cpe/p3 tuners who don't know about di???? - if i'm wrong prove me wrong - somebody's gotta know what's going on there with all the rumors of what' s to come...

when you show the average tuner around the sb (which i have done locally) - they think in terms of only what a piggy back can do, and they probably don't think in terms of the complex di system on the ms3 - they prob think in terms of what piggy backs do on port injected cars - which means stick to tricking the maf for fuel and tricking the cps/csps for timing, add fuel & if you get knock pull timing, and if you need more fuel just get bigger injectors - well the rules change with di - the injectors are already supposed to push tons of fuel - so then you have look at the complicated, most likely load based equation that is involved in fuel pressure and who really knows how to do that? why did lou put the table there if jordan didn't know how to use it?

look this idea to me is like using a maf clamp instead of tuning - you can get the job done, but it's like a broad sword when you need a razor -

no disrespect to dj here - he's way smarter then me to have come up with this, but i gotta say you have to be able to get further with the sb dialing in exactly what you need

granted this is some serious speculation too, but when all the factors are presented to me, these are the conclusions i draw...

You'll get no disagreement from me, the sb definitely allows for much more fine tuning of the pressure than the "trick" i've come up with. But IMO, with the ap and a good tune, you can correctly account for the fueling offset without issue. Think of it kinda like bigger injectors. Are bigger injectors a "broad sword"? Sure. With "upped" pressure, our stock injectors will behave exactly like bigger injectors (hopefully without any problems, we'll see i guess, lol). 22% more flow across the board, theoretically. Bigger injectors require a re-tune, and so will this mod.

Part of me almost thinks that having the capability to vary the pressure at different load points, or rpm or whatever, only complicates this approach. Sometimes simpler is better, imo.

sb3?
i think what's talked about here will prove to be superior to anything - the only other thing that could be as good is the mazdaedit software discussed here:
http://www.mazdaspeedforum.org/forum/foru...ailable-38714/
nobody is interested in this though... so many are cobb fanboys

i have been researching cars in the hopes i will move to another platform - i think the real secret to the cobalt reaching over 500 is these guys
HPTuners.com >> Performance At Your Fingertips

piggybacks or flash devices are great - but i just don't see anything better then getting in there and really re-writing existing ecu tables - besides of course running all the sensors from a standalone

Yeah it would certainly be ideal if Christian / Cobb could find the parameter for the fuel pressure so we dont have to dance around it. But i think DJ's stuff will work just fine if tuned for.

I was VERY interested in that software I even committed to spend almost $1000 for his pro version and the cable. All I asked of him was a screenshot showing his software disabling the EGR low flow MIL.. That was the primary reason for me even considering his software--and he had stated specifically to me that it could delete that MIL.

He could not produce a screenshot, and then stopped responding to my PM's on his forum...

cpolly and myself talked to him in person, I was all for doing it as well, but he wants a bone stock car to work with, I guess so he can see what can be done with it before any bolt-ons are added

Trifecta gets it done: Fuel tables unlocked - Cobalt SS Network

interesting. ha

I cant believe im thinking of getting a cobalt...theres a guy on ther at almost 360 hp w/ dp,intercooler, and a tune at 22psi...

Ohh and one more thing. Matt @ Zzp said "we are considering working with the mazdaspeed platform"