DIYEFI.org Forum

I've taken some slack for my addition of $.50 in parts on freeEMS 1.0 for current sensing capabilities on the driving circuits. As luck would have it, relative to my wife's 2001 VW Jetta, I really wished I had on board current sensing capabilities. The lack there of, didn't do much more the shorten the life of the turbo, but still quite annoying. Some history.

About a month ago, we noticed a, intermittent whooshing sound when we spun up the turbo. You know the sound from the blow off. Which would indicate the waste gate might not be working correctly. However the waste gate seemed to move freely, ect. I couldn't find a finger to point.

About 4 days ago, she called and told me it was bucking and she was concerned it wouldn't make it home, but it did. So I took a look at it when I got home. Seemed a cyl wasn't working for one reason or another.

Today, I did a compression test, 145 psi in all 4 cyls, cyl 3 had a wet plug and all cyls had a light brown color. Indicating it was running well, then suddenly stopped firing. So I did the spark test, and low and behold, it wasn't sparking. Further testing showed it's the coil pack. It sits on the plug, so now spark plug wires, ect. Easy to replace, once I get my hands on the parts.

I'm certain the issue I was seeing prior to the catastrophic failure was a multi layered intermittent problem. I found that the second I touched the system, it would start working. So no problems right.

I'm certain that what was happening was that the cyl was not firing completely, causing extra fuel to be dumped in the exhaust. Spinning Mr. Turbo beyond the waste gates control capabilities. Resulting in the blow off's activation, and over RPM of the turbo.

Unfortunately, the oil consumption has increased from what was .25 qt per change, to 1 qt per change. Quite an increase in consumption in just one oil change. I also see some oil dripping from the air cooler. After seeing the compression test is dead on, I have to conclude it's coming from the turbo. Another indication that it was over spinning.

If I had the ability to see the current going into the coil pack, then compare it against the normal current, I'm sure I could draw differences that would indicate failing components. So as far as I'm concerned, it's a good $.50 to invest.

I know that it won't be supported in the first rounds of software, and I plan on having the PCB's created with jumpers installed similar to the included fuse theory. If you want the feature, or some day we add the required software components, then you can scratch the jumper off, and add the $.50 in parts later.

Any how just wanted to share a bit of a story where I'm sure it would have helped diagnose an intermittent problem.

jharvey wrote:I've taken some slack for my addition of $.50 in parts on freeEMS 1.0 for current sensing capabilities on the driving circuits.

It's certainly not the parts that I care about, it's not even really the extra circuit complexity, what it is is the insertion of what I see as unrequired parts into critical paths from sensor inputs and output drives to the world.

About a month ago, we noticed a, intermittent whooshing sound when we spun up the turbo. You know the sound from the blow off. Which would indicate the waste gate might not be working correctly.

There are two things seriously wrong here. Possibly by manufacturer design. 1) the BOV should never leak unless there is vacuum signal to it or the boost is WAY high. The bosch BOVs used on most turbo euros are absolute rubbish and function with a diaphragm over a hole that is just asking for trouble. Get and fit a DSM/evo one to it and be much happier, they won't leak until the boost gets rather high. 2) no amount of ex flow short of antilag should be able to circumvent the gate action. The gate is controlled by intake pressure though, so if it never goes high enough it will never fully open. Hence 2) is caused by 1)

Further testing showed it's the coil pack. It sits on the plug, so now spark plug wires, ect.

What about the coil pack? Did the carbon rod crumble/burn out? Of the actual coil has failed? Or does it have an integral ignitor that has failed?

I'm certain that what was happening was that the cyl was not firing completely, causing extra fuel to be dumped in the exhaust. Spinning Mr. Turbo beyond the waste gates control capabilities. Resulting in the blow off's activation, and over RPM of the turbo.

Firstly, not firing/firing late doesn't mean more fuel, just less energy into pushing the car and more into the manifold. Again, the gate can only activate properly with an appropriate pressure signal. Those bosch valves :



are junk and troublesome at stock boost in the first place. If the BOV had stayed shut, the pressure would have risen causing the exhaust pressure to rise and the actuation pressure on the gate to rise and ensuring that the turbo didn't spin freely with no compressor load.

I also see some oil dripping from the air cooler. After seeing the compression test is dead on, I have to conclude it's coming from the turbo. Another indication that it was over spinning.

How can oil drip from the air cooler? air cooler = intercooler? If so, that is a sealed system. If it isn't sealed, that is another problem causing your turbo to spin faster and work harder than it otherwise would have to.

If I had the ability to see the current going into the coil pack, then compare it against the normal current, I'm sure I could draw differences that would indicate failing components.

Firstly, you are assuming no spark to the plug means no spark. Secondly you are assuming the ignition is driven from inside the ECU which 99% chance it isn't. Thirdly your design has a digital input which says "current or not" and doesn't give you a contour of current. If you want to get the length of the current at some threshold level, you are out of luck for the same reason we can't do v8 sequential properly; not enough timer channels. What you want is ion sensing and misfire detection etc. These are both things not done in the ECU but done outside and passed back in with a digital signal in OEM apps anyway.

I plan on having the PCB's created with jumpers installed similar to the included fuse theory. If you want the feature, or some day we add the required software components, then you can scratch the jumper off, and add the $.50 in parts later.

You should have said that earlier. I have much less problem if the average user doesn't have to install an extra jumper in a critical location during their build. Provided the presence of locations for these parts doesn't interfere with the layout in a negative way, it doesn't matter to me :-)

Fred.

Fred wrote:What about the coil pack? Did the carbon rod crumble/burn out? Of the actual coil has failed? Or does it have an integral ignitor that has failed?

Good question, I really don't know how those coils work. My assumption is that low power level signal is run to the coil, then the power driver (FET, IGBT, what ever) and coil are packed into one unit that snapps to the top of the plug. When you say carbon rod crumble / burn out, I'm assuming you not talking about the spark plug. Plugs were in great shape, almost no errosion at all, gapped correctly ect. Other than that I haven't been able to test them further. I also know they have 4 wires, I'm not sure what the 4th is needed for.

How can oil drip from the air cooler? air cooler = intercooler? If so, that is a sealed system. If it isn't sealed, that is another problem causing your turbo to spin faster and work harder than it otherwise would have to.

It was a very small leak, small enought that I'm sure any oil in the hole, blocked air from passing. I sprayed with soapy water, but couldn't find a leak. There is also a chance it wasn't a leak, but I don't have an different explanation for how oil got there. Yes intercooler, I had a brain fart when I wrote that.

I plan on having the PCB's created with jumpers installed similar to the included fuse theory. If you want the feature, or some day we add the required software components, then you can scratch the jumper off, and add the $.50 in parts later.

You should have said that earlier. I have much less problem if the average user doesn't have to install an extra jumper in a critical location during their build. Provided the presence of locations for these parts doesn't interfere with the layout in a negative way, it doesn't matter to me

That one had just occured to me this weekend, when I was looking into footprints. So I let you know not long after it popped in my head.

jharvey wrote:My assumption is that low power level signal is run to the coil, then the power driver (FET, IGBT, what ever) and coil are packed into one unit that snapps to the top of the plug.

With 4 pins I'd say you are definitely right :-)

When you say carbon rod crumble / burn out, I'm assuming you not talking about the spark plug.

Correct, typically they have a carbon conductor between the coil and plug much like a lead has inside of it. These can give trouble but typically only in older COP cars.

Plugs were in great shape, almost no errosion at all, gapped correctly ect.

I hope by "gapped" you mean "not touched at all" because no modern engine I know of needs it's plugged gaps at all. On the contrary, you specifically should NOT gap any modern engines plugs. If you are making so much power that you are getting a high load miss then you need a better coil or more dwell etc, not gapping.

I also know they have 4 wires, I'm not sure what the 4th is needed for.

I did know this, but I've forgotten again.

That one had just occured to me this weekend, when I was looking into footprints. So I let you know not long after it popped in my head.

OK, cool :-) I was thinking it, but didn't want to suggest it in case I pissed you off :-)

Fred.

I had the same symptoms when my Nissan 200SX turbo started to die.
The oil bearing on the standard turbo got knackered and started making a shrilling noise under boost.
the worn bearing also started spewing loads of oil into the inter cooler which would intermittently bubble and blow into the engine under high boost. this would cause a massive plume of blue/white smoke out of the exhaust and serious misfires, I lost cylinders 1,3 and 4 in the same week all caused by oil fouling on the plugs. Once cleaned it would go fine until you put a little boost on and push the oil up from the inter cooler again.
When i got around to investigating what was causing it there was over a pint of oil in the inter cooler, the turbo had half a mm of lateral play in the bearing and the cat was shot.
All of the pipes both intake and exhaust had loads of oily grime in them which needed cleaning out.
after fitting a newer turbo and cleaning it all out it was perfectly reliable until i turned the boost up to 14psi and killed the bottom end.

So in short, check your turbo dude!