Electronic VNT controller design - Updated!! Rev 0.1 :)

[Vincent Waldon]

And even though I don't speak polish I can translate the schematic... interesting design.

About 50 bucks in off-the-shelf parts... pressure transducer reads manifold pressure as an input to a PWM controller driving the ubiquitous N75 vacuum valve as a vacuum-operated vane controller.

No switch for "vane position limit" however.    :wink:

[gigaz2]

that device ultimately only translates boost into vacuum, nothing else

boost rises, pressure sender voltage rises, pwm rises, N75 makes vacuum lower, that moves the vanes, boost lowers, cycle repeats

its like swapping the vacuum can for a boost one, but with more adjustability

"our" device would be basically like that, but with pedal input as well.

its a good starting point nonetheless, but pwm is made by the microcontroller witch is very cheap.

[giulianot]

Finally we have got the ball rolling, let me know is you need any stock vnt 15 vag graphs for some baseline info.

[Vincent Waldon]

The RPM input is a tricky one, as being old school diesel, the only way really will be from the W terminal of the alternator.  I have done some research, but found no scope outputs or verifiable documention.

Yup, on my tired old analog scope it looks like this:



Schmitt trigger or a zener should clean it up nicely.  Or a PLL.

[Sprockets]

Vincent: Many thanks for that scope trace, it's exactly the sort of info I need.  I now feel confident enough to get on with the main schematics, and mock up a prototype hardware circuit.

I will also have to knock up a test bench aswell, will rummage around the spares for an alternator and vac pump.  Throw in a couple of vac guages and we should be set for some code!!  The one thing I'm not 100% looking forward to is the front end, but thats a while off yet.  I'll just edit the values in code myself for a bit  Was thinking of a QT based GUI so as to be cross platform, or even java.....but I'll cross that road when i get to it

You prolly won't hear from me for a bit now, got other work lined up, but next post should be with the Eagle schematics!

-Gavin

[zaprzal]

Can anyone translate this?

http://www.forum.vwgolf.pl/viewtopic.php?t=147110&postdays=0&postorder=asc&start=0

Andrew

Ventil posted it here:
http://www.vwdiesel.net/phpBB/viewtopic.php?t=11793&postdays=0&postorder=asc&start=15&sid=fa1c7230b92b8eb7b1543320f54a9223

Ahh, thanks.  I knew Alta Vista, but they don't do Polish.

Andrew

http://translate.google.com/translate_t#

by the way online translators sucks

[gigaz2]

that signal already seems clipped to me, I'll try to figure it out on my open alternator

[Sprockets]

First post updated guys.

Ideas/criticisms/telling me to go away and do summat useful with my life etc all welcome

-Gavin

[gnavs]

I absolutely love the effort you're putting into this, I just wish I understood more about electrical design in general  :oops:

[regcheeseman]

As suggested, ignore the w term altogether. It cannot serve as a datum, if prone to noise and shift and varies with differing market versions of varous models.

I'd either go for a proxy sensor onto flywheel through bell housing or a tooth count or air gap pickup from the cam pulley.

I've built a prototype mapable gasser ignition for my dirt bike, and have considered a diesel variation but it's a case of too little time, too many projects...good luck

[CoolAirVw]

vwcaddy built a photo optic rpm sensor that he added to the timing cover reading off the injection pump pulley.  

http://www.4crawler.com/Diesel/CheapTricks/Tachometer/index.shtml#PartsList

[blackdogvan]

Just bought a LR 300 pump for M-TDi build.

Has perfect throttle position sensor integrated directly onto throttle linkage of pump. Looks to me like a simple weather-proof rotary pot.

Though of you guys when I yanked it off.

[Vincent Waldon]

Pictures please !!  Rotary pot integrated into throttle very cool !!

[Sprockets]

Regarding the RPM input query:

If I was having to drive a highly precise time driven application (ie, ignition system on a petrol engine, or the injection of a diesel engine) then yes, the W terminal is a crap method, and frankly pointless! But for this application, it's a perfectly viable solution.  The rev counter on your old diesel works fine from it doesn't it?  Granted, it's not accurate to the nearest 10rpm, but do you need to know that sort of resolution? No.  same in this case.  As long as I can get consistent readings, which I believe with a well maintained car (read, alternator belt in good condition lol) I can, it will be an acceptable solution.  Also, if you personally feel that you want an RPM signal that will be precise enough to show you the compression waveform of the engine, the hardware will handle it just fine, just needs the external stuff bolted to the engine.  But as I've always said, I don't want people to have to do that, just give them an option. The RPM is configurable in software for the offset and ratio multiplier.  Once I get a database of pulley sizes, it will be easy enough to set the 2 variables, or in the end, just but a little calcualator in the front end gui.

And as for the throttle pot, yep, good find on the 300TDi, but you will find most mid to late 90's diesel stuff (well, here in the UK anyway) have this sort of stuff to handle all the emissions stuff that had to get bolted onto em to try and clean em up!

Anyway, sorry for my rambling with the RPM signal, just had a brain dump to my keyboard!

Cheers

-Gavin

[giulianot]

hello Gavin, I think your initial design looks great . Have you given any thought to what programing language you will use . I was just wondering is some stock tdi data logging would be of any use to writing algorithms.