Hi,

My friend recently bought a dirt cheap turbo 1.6l 2000 NB Mx5. The previous owner had paid some shop to turbo it back in 2003 or smth and had no idea what the shop actually did to the car. He was the only owner of the car and drove it 130000 miles before selling it. Obviously when my friend bought the car it was in "perfect" condition with no problems whatsoever. However couple of days after bying the car (no driving between buying it) it developed an no-power issue where it sometimes completely dies when you slow down to stop, and it is really struggling to even move with below 3k rpms. The car idles barely but smells fuely as hell. On low rpms you can also clearly feel that it is not igniting if you give it a lot of pedal. Kinda stumbles as if the there would be a coilpack issue.

Im thinking that it could be a either indeed a coil related issue, or something to do with the fuel mixture like broken sensor or smth. I was curious enough to pop open the ECU to see if there is anything obviously wrong (we had no clue if the car has an proper aftermarket ECU or a stock one), and to me it looks like there is some kind of a piggy-bag chip on the stock ECU to make the car run with the turbo setup. On this chip it looks like there is a resistor that seems to be a bit dark on one end, indicating that maybe it has been shorted or smth?

I suggested my friend to by a cheap coilpack from ebay to rule that out, but other than that I don't really know which direction to start going. However I would like to know if anyone recognizes this chip, and opinions if the red resistor looks normal or not? As I think one option could be to try to find this chip from somewhere (or just soldering a new resistor), or just buying a proper ECU straight-away and ditching this crap incase the chip has actually gone bad.

I did not really have time to properly look at the car, but I think the lambda also was not stock, and I don't remember seeing the black OEM MAF-tube thingy anywhere (could just be relocated into a weird place).

Fun stuff. Sorry for bad photos, operation was done slightly drunk.

 

I do not see any damaged components on both of those PCBs.
The modification to this board has only 3 wires, so I expect it cannot do much.
To me it seems most likely this modification alters the coolant temperature reading or the MAF reading of the ECU, causing it to enrich the AFR.
I also expect the stock MAF to be retained as I cannot think of a way to let the stock ECU run the engine without MAF.

If you want to run this engine properly, use a different ECU solution. If you are tech savvy and do not mind doing a bit of tinkering/development, there is a Speeduino NB PnP ECU in the making (I think you can contact @noisymime if you want to beta test one) and that would be your cheapest option.
There are also other PnP options from MSlabs, DIYautotune and some UK firm whose name should not be mentioned.

 

My thoughts exatly, I am not a super duper expert on boosting Miatas but I was pretty sure this is the case.

Okay, thanks for the tips. Good point about the wiring, I think I have to look a bit more into the car and try to follow the wiring from the lambda or smth to make sure I did not miss anything relevant about the setup. Definitely the ECU should be updated anyways I think, maybe I will consider contacting this guy you linked if I am brave enough for some proper diy, since this friend of mine for sure will like the sound of affordable But I do have some friends and I know many people running Reverants MS with good experience too.

The guy who sold the car actually said that he had himself somehow replaced the "on/off" oil-pressure indication with a real reading. I am not sure exactly how he did it, if he replaced the sensor or what, but I am kinda wondering now if this chip here instead could have something to do with that..

 

The oil pressure sender has nothing to do with the ECU. On an NB, it is simply an on/off pressure switch directly connected to the dial in the dash.
I will point you to this excellent post from Joe Perez with links to modding the NB dial: https://www.miataturbo.net/general-m...87/#post988475

 

It looks like this board intercepts the stock ECU's flash memory. With only three wires, I doubt it can do much, but it will be interesting to analyze it a bit more. See if you can take better pics.

 

You are right, I have to rectify my post.
One of the chips is a 74HCT132 which is a NAND schmitt trigger, so a digital IC.
The pink wire is VCC and the brown wire is GND. It looks like the green wire goes into one of the DQ* pins of the flash memory.
Maybe some device to flip the MSB of the address bit of the flash to run an alternative map?

 

If it's a schmitt trigger, it's probably some way of better filtering the NB cam signal.

 

I do not think the cam input is related to the flash chip.

 

Yeah, curious... Was thinking that was tied into an interrupt. I blame lack of caffeine.

 

Just a quick update for anyone curious, tried swapping the coilpack, did nothing. I also found the MAF and tried swapping the MAF and and IAT sensors from my car to his car -> no change.

Next up would be cam and crank sensors. If these won't help I think I'll check the lambda, if even that does not help, I will just bin the ECU and we put an order for a proper one I think. It's a slow process as I am busy as hell and have only so much time to help him.

Funny thing I found out, that this company that installed the turbo-setup, put the intercooler UNDER the radiator. So it is basically pointing straight into the engine cover on the bottom, or somewhere inside the bumper. No holes no nothing have been cut to the cover or the plastics inside the mouth of the Miata, meaning that the intercooler propably does absolutely nothing but increases the price they could charge from the guy who paid for it. The pipe you see going down in the pictures is going to the air filter which has been moved to somewhere infront of the right front wheel.