joyrider

iTrader: (8)

Car is a 2000 turbo BEGI/AVO, GT2560, no AC, PS, stock injector atm and stock IAT.

I've redone my alternator circuit regulator with the good parts ordered from mouser. Checked the circuit and it's good up to 14.48V until the field output drop.

Test it on the car and I have few problems, bad idle, inconsistent AFR and few other minor issues.

Took the 12V input of the circuit off and retried it. Idle, AFR and everything seems fine running only on the battery.

Here's 2 logs, first with alternator control and the other without.

WestfieldMX5

Does it charge the battery when hooked back up? What voltage with engine idling?
Is battery voltage stable?

joyrider

iTrader: (8)

Yes it does, battery is at 12.5V and 16.4V at idle.

I found that similar problem we're discovered by people running NA alternator on the NB. I will check wich alternator is in the car as I'm the second owner. I will put the OEM Ecu and take voltage mesurement at the battery too.

I'll let you know what I'll find.

Braineack

iTrader: (62)

thats what happens. output turns off when input voltage goes above ~14.Xv as soon as input voltage go back below, the output comes back on.

joyrider

iTrader: (8)

Okay here's the news, still a Hitachi BP4W in the car.

OEM Ecu : battery idle is 15V, 15.5V max with some throttle.

Looked at

Did change a resistance (48.7K) and now the voltage at idle is 16V, 16.5V max with some throttle.

Do know if it's too much but idle was better than previously.

WestfieldMX5

I hope you meant 14.6, not 16.4V
I have built a couple of these circuits and noticed different behavior. Some circuits have very stable voltage (flat curve at idle), other circuits were a bit less stable (0.2V variation).

Edit: 16V is too high. You should aim for 14.4-14.6V
Your bulbs, battery and electronics will be grateful.

joyrider

iTrader: (8)

It was 16.4V.

Don't know if I could cheat the resistances to have a better result. The strange thing is I tested my circuit prior to reinstalling the DIYPNP in the car and it was fine following 12V up to 14.48V and dropping after.

Do you know what the math behing this circuit ? I could try to cheat a little.

Braineack

iTrader: (62)

when i do that same circuit it switches at 14.2v, using all 1% resistors. there is some post where it was discussed, maybe back in 07-08. joe and jason talked about changing the 48.7 to move the voltage around. it's important you use 1% resistors otherwise, the 5% ones can throw it off too much.

does the voltage as displayed by the MS match the voltage at your injectors from a DMM? it's important the DIYPNP is calibrated to see the same voltage as the injectors.

joyrider

iTrader: (8)

Yes, I remember seeing some math around this circuit but can't find them.Joe Perz if I remeber right... On my bench the circuit switches to 14.48V and yes I used 1% all around. I will try to take reading directly on the circuit while the car run, PITA with the ECU location tho.

DMM is about 1V less then the battery poles wich seems to be okay.

Braineack

iTrader: (62)

but make sure the voltage at injectors, is exactly what the MS is displaying.

joyrider

iTrader: (8)

Found the math behind the circuit : Max Output = (1+48.7K/10K) x2.45 = 14.38V in theory.

Went to take measures on the board itself, 15.9V instead of 12V but 0.5V to 1O field output so the circuit is working.

I think my alternator is done, will try another one and let you know. I will check the voltage at the injectors too, thx for the tip.

joyrider

iTrader: (8)

Alternator was tested and is okay. Seems like only a ground issue is left.

Ground on 4a and 4b on the diypnp, dont know whats missing.

Braineack

iTrader: (62)

and 4c and 4f are grounded too right? at least through SG?

got a pic of your circuit, top and bottom? maybe we can spot something.

joyrider

iTrader: (8)

Yeah 4c and 4f are SG. Will take a pic or 2 tomorrow.

Voltage went 14.6V with SPAL fans on, lights and defrost. That can be a quick fix for the next track day. Every grounds, and the one added directly on the alternator frame are good.

joyrider

iTrader: (8)

Just to let you guys know, sent the car for alignment and got the car and voltage was fine. Ask my mecanic for what he found and he didn't touch anything. Strange but who care it works.

Went to TA yesterday and during session, the trouble comes again and messed the car up + heat soak didn't help while autotunning in practice session (lesson learnerd the hard way). And the worst of all, the damn ECU fell off (under the dash) and blocked the brake pedal at the end of my only good session. Brake we're stuck on and the car was really hard to move and CLT went to the sky but I had to box to keep my time. Boiling prestone/water everywhere on the grid where the car stopped. The reserve tank just burned up and spill everything, log shows 293* for max CLT.

Time to get this thing fixed, change the brakes and look at the engine if some harm was been done. I would like to hook up my EGT reading to my DIYPNP too, will look for that also.

BTW I took Brain's AFR map as base map but I only run 7psi (158KPa), wondering if I should go richer then your tune in the 150-160 KPa range and maybe everywhere since it's a dedicated track car.

Braineack

iTrader: (62)

I'd go 11.8:1 everywhere in boost on a track car.



FWIW, I just sent the alt circuit to get printed on a pcb layout i whipped up.

joyrider

iTrader: (8)

Could be better this way, EGT's seems to be way to high. At least it was a short track.

joyrider

iTrader: (8)

Okay back on voltage regulator issue. So alternator tested another time by someone else ans it's good. Show him my circuit and it send voltage up to 14.4V so that's good too. He told me to check the wiring. Everything seems fine.

One question tho, while browsing the 2000 wiring diagram, I see both wires to the alt (gray and gray/red) are going in the PCM (1T and 1O respectively). My field output is connected to 1O and when looking at the diagram, it seems this wire should receive while 1T transmit. Here's the diagram page B-1c.

http://www.madracki.com/miata/images...g/2000wire.pdf