need help with Emerald ECU issue

[rzempel]

I'm tuning my performance built 2.0l Zetec, with Emerald K3 ECU, and have run into a problem that I can't figure out. When it's warm, it starts, idles, and runs great. But, when it's cold it will not start unless I reload the configuration file and re-calibrate the throttle sensor. That, to me, makes no sense whatsoever, so I must be missing something simple.

Does anyone have an idea what's going on?

Randall

[magnusfeuer]

I don't know the emerald that well. But having worked a bit with embedded systems (which the emerald is), I'd say you have a bad battery inside the unit.

The battery is able to hold the stored configuration for a few hours, but is too weak to keep the CMOS ram (or whatever they use) alive over night.

Disassemble the unit and see if you can find a button battery. Replace it and try again.

/Magnus F.

[rzempel]

That's an interesting thought. I'll check it. But, we just tried the reload, and it didn't work this time.

[magnusfeuer]

Hmm.

Is there a way to "restore to factory configuration"? A bad battery can not only forget configuration, but also scramble it to the point that the configuration software gets confused. Please note that I am speculating here.

/Magnus F.

[Doug Liedblad]

What does Emerald say about the problem?

I think they have some distributors in the US.

[GWise]

Sounds more like a problem with the throttle (or possibly some temperature) sensor than the ECU if it's a temperature thing. What happens if you recalibrate, start the motor, shut it off, and try to start it again before it's had a chance to really warm up?

I had a temperature sensor that would work fine when it was a warm, but below a certain temp I had no reading at all...

[rzempel]

Dave Walker at Emerald said, "[relaoding the configuration file]...makes no difference to starting." He has now gone home for the weekend, so we're on our own.

When we reload the config file, it is to "factory condition"--that's the file we have.

[slomove]

I don't know the emerald that well. But having worked a bit with embedded systems (which the emerald is), I'd say you have a bad battery inside the unit.

The battery is able to hold the stored configuration for a few hours, but is too weak to keep the CMOS ram (or whatever they use) alive over night.

Disassemble the unit and see if you can find a button battery. Replace it and try again.

/Magnus F.

I am not aware that there is a battery the Emerald. At least the manual does not mention it and if so it better be a lifetime battery.

Most such systems nowadays use flash memory instead of CMOS RAM.

I would also go with a sensor problem if there was not that funny effect that reloading the .fig file would make a some kind of difference. For that matter, the sensor calibration info is not even part of the configuration file.

If you look in the "life" tab window, you can see all the sensor measurements, i.e. air temp, water temp, baro and TPS. I would check that they show reasonable values when the cold start problem occurs.

Gert

[rzempel]

Everything is normal in the live values window. Tha's part of what is so maddening about this.

[moosetestbestanden]

Just go get a can of ether heptane like any Lotus owner would and give'er a quick squirt. That ought to do it.:D

[Doug Liedblad]

Randall

Did you find the problem?

[rzempel]

Not yet. We are going back through everything...battery condition, wiring, sensors, etc.

[magnusfeuer]

Transitioning from other post.

When you cold start with a messed up unit, I take it that you get a TPS range that differs in comparison with after you recalibrate it. Is this correct?

Also if you compare the map you have at cold start, is it identical to that of the reloaded map?

Lastly: Where are you located? If you live in SoCal and can drop the car of in my garage and let it sit overnight, I'd be happy to have a look at it.

/Magnus F.

[rzempel]

When you cold start with a messed up unit, I take it that you get a TPS range that differs in comparison with after you recalibrate it. Is this correct?

I'm not sure what you're asking. Once calibrated, the TPS readings are consistent.

Also if you compare the map you have at cold start, is it identical to that of the reloaded map?

It should be, shouldn't it? As I understand it, I have to make deliberate changes to the maps--they shouldn't change on their own. That said, This morning I did compare the map in the ECU with a "fresh" copy of the file. The one in the ECU looked like it was written by a random number generator. I saved that one for later head-scratching and loaded the clean file. I added a bit of fuel, and the car started right up. I'm leaving it sit for a bit while I do some errands. In a bit, I'll go back and see what happens.

On the dyno, the car would start easily when the ECU was warmed by a hot-air gun, but not otherwise. When the engine and, presumably the ECU, was warm, the car started easily.

I drove the car from the shop last Friday. It ran great, but became increasingly reluctant to start even when warm. As I thought about the symptoms, another possibility occurred to me.... The tuner was using ether to start the car from cold--sprayed onto the filter gauze. The filter is immediately behind the ECU. What if the filter was still saturated with ether, and warming the ECU was in reality evaporating the remaining ether? I tested that idea this morning, and voila, the car started nicely with just a bit more cold fuel enrichment. Maybe that's all it needed.

Lastly: Where are you located? If you live in SoCal and can drop the car of in my garage and let it sit overnight, I'd be happy to have a look at it.

Yes, I'm in SoCal--La Habra Heights, near the borders of LA, Orange, and Riverside counties. Where are you, in case I haven't got things figured out after all...?

[slomove]

........On the dyno, the car would start easily when the ECU was warmed by a hot-air gun, but not otherwise...........

I would bet a buck or two that your ECU has some cold solder joint, other temperature-sensitive intermittent or some damaged chip. Especially if the memory is scrambled without good reason.