Show full size 924Board.org
Discussion Forum of 924.org
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 
 Technical FAQ924 FAQ (Technical)   Technical924 Technical Section   Jump to 924.org924.org   Jump to PCA 924 Registry924 Registry

931.606.021.00 Pulse sender, flywheel sensor? 1981 924 Turbo
Goto page Previous  1, 2, 3, 4, 5  Next
 
Post new topic   Reply to topic    924Board.org Forum Index -> 931 Tech.
View previous topic :: View next topic  
Author Message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Tue Apr 24, 2012 8:03 am    Post subject: Reply with quote

Does anyone have a schematic started?
All the chips I have.... mostly...
But the chips are digital logic. The timing must be done by the caps.

If I had a schematic I could see where exactly it happens.
Box caps are good... usually solid. Don't drift over time.
Back to top
View user's profile Send private message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Thu Apr 26, 2012 1:58 am    Post subject: Reply with quote

I've tested this crank sensor in the car a few times, and I'm not getting any signal. Measuring in DC, AC volts... and even using my scope.
(With engine cranking)

I feel confident that I need one of these crank/flywheel sensors.
No sources. I can't find one.
Anyone got pointers?
Back to top
View user's profile Send private message
ideola  



Joined: 01 Oct 2004
Posts: 15510
Location: Woodstock IL

PostPosted: Thu Apr 26, 2012 2:19 am    Post subject: Reply with quote

As noted, they are NLA. You will not find a new one, I can pretty much guarantee you that. Your only option is to find one from a 1981-82 931 that is being parted, or to try to hunt down the BMW variant that is rumored to work (or some other alternative).

I don't know much about VR or Hall sensors, but I'm wondering if it has to be energized to generate a signal? I find it strange that your readings are identical to three separate sensors that I've tested (one IN a car, the other two known good spares removed from running cars).
_________________
75 Fiat X1/9 | 80 US 931 '941' | 78 Poli-Form
Back to top
View user's profile Send private message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Thu Apr 26, 2012 3:04 am    Post subject: Reply with quote

I'm searching porsche clubs and parts locations for this sensor.

I wish this was a Hall Effect based sensor. Then it would be easily replaced. A VR sensor is basically a magnet sitting in a coil. As the flywheel moves past the sensor, the magnet moves in/out of the coil and thus creating a volatge. (Same idea as those shaker flash lights? Seen them?)

Anyways....
A VR sensor does not require an external voltage, but creates one. AC voltage. This is seen by the DITC.

I suspect the magnet in mine is stuck...not moving. The resistance we are measuring is just the coil.
Back to top
View user's profile Send private message
ptheskil  



Joined: 03 Aug 2004
Posts: 128
Location: Essex, UK

PostPosted: Thu Apr 26, 2012 9:46 am    Post subject: Reply with quote

Matilda wrote:
A VR sensor is basically a magnet sitting in a coil. As the flywheel moves past the sensor, the magnet moves in/out of the coil and thus creating a volatge.

I suspect the magnet in mine is stuck...not moving.


Not quite. Simple VR sensors do have a magnet in them but it is static. It produces a magnetic flux which is coupled to whatever ferromagnetic material gets near it. As the teeth on the flywheel pass the sensor the magnitude of the flux changes which gives rise to the AC voltage as you correctly state. This voltage is a function of rate of change of flux - meaning that the faster the flywheel moves the higher the voltage. At crank you won't be getting much and at 6000rpm it could be in the 100s of volts.

If the continuity/resistance between the pins checks out there's not much can go wrong with it. However, if it is not positioned so that there is a very narrow gap between the teeth and the sensor face you will not generate enough voltage at crank for the DITC to make sense of it. You should try and measure the voltage across as high a resistance as you can so using the input of a DVM on VAC setting is all you need to do. As you have a scope I would expect you to find something though even if it is a very low voltage signal. Check that you are not shorting the leads from the sensor or placing a low resistance across them as that will destroy the signal.

Some time ago when I had problems with my ignition system (turned out to be the DITC) I mapped out the pins and what they were for. My old notes tell me that pin15 is a ground for the shielded cable (which is not connected to ground at the business end) and the CPS signal is on pins 13 and 14. Assuming this is right (!) you should be looking for a signal between pins 13 and 14. Checking the resistance between the DITC pins you should see: pin15 to pin12 (ground) should be close to 0ohms, pin13 to pin12 very high resistance possibly open circuit depending on your DVM, pin14 to pin12 around 15megaohms.

Can you check the positioning of the sensor in relation to the flywheel to check the gap width?
_________________
1981 931 series2 Euro spec
Back to top
View user's profile Send private message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Thu Apr 26, 2012 10:10 am    Post subject: Reply with quote

I've tried using my DVM (which has a high impediance input) across the CPS terminals 13/14 of the flywheel sensor, while I was cranking the engine. No voltage seen. I also scoped pin 13, and then 14 (wrt ground) while it was plugged into the DITC, while cranking the engine.
The DITC was getting power & ground.
I confirmed that the DITC inputs (At this sensor) were not internally shorted to ground. They show lots of resistance. Mostly because of the two box caps that are in series with these lines, to the rest of the circuit.

Years ago I checked a VRS sensor on a VW using a similar meter, and I remember the meter showing something like 2.5V AC (At engine idle).... So I feel confident I'm testing the sensor right.

I also tested the crank sensor by puting a 2.7KΩ across the cps terminals, with my meter in parallel to that. Nothing...

My questions are:
-How would I loose gap?
-What should the gap be??
-Would you agree that it's possible to have the right resistance values, but still have a faulty sensor?
Back to top
View user's profile Send private message
ptheskil  



Joined: 03 Aug 2004
Posts: 128
Location: Essex, UK

PostPosted: Thu Apr 26, 2012 8:52 pm    Post subject: Reply with quote

OK so a few more thoughts...

Engine speed at crank is say 150rpm vs say 1000prm at idle. 1-2V-ish sounds about right for a CPS signal at idle but it will be much lower than this during crank. I wouldn't be surprised if a DVM (even a good one) struggled with that as the signal will be far from a regular sine wave during crank as the engine speed varies widely due to compression events. I think you said you have access to a scope; try and take a look at it on that. If there's anything there you should see it with a scope for sure. Be careful with the probe grounding and obviously do the test with the sensor disconnected from the DITC.

Anything is possible of course but it is highly unlikely that there will be a problem with the sensor if the resistances check out and all seems well with yours. Not impossible though so we should be open minded.

In terms of the gap it should be as close as possible - no more than a mil or so. That's about 1/32 inch in the old English you use over there.

Is it possible that the sensor may have been knocked or disturbed on the engine? The other possibility is that it got chewed up by the teeth but I would not expect the resistances to check out in that case. Not impossible though as stated before.

It's starting to sound like swapping with a known good DITC is the next course of action. At least then you will have another piece of evidence to justify replacing the sensor. You really don't want to have to do that.

My apologies if I am telling you things you already know - just trying to be thorough with the approach here as you seem to have a real puzzle on your hands. Good luck.
_________________
1981 931 series2 Euro spec
Back to top
View user's profile Send private message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Thu Apr 26, 2012 11:40 pm    Post subject: Reply with quote

All advice is good.
I understand that the voltage, at crank, would be low from the sensor. But I've scoped this sensor when it's plugged into the DITC. I'm scoping all 3 lines. Nothing. Even with the scope set to 0.1V per division.

So yes, even with the voltage at crank being low, there still must be something there. Something the DITC can see. Something my scope should be seeing!

I think what I'm going to do next is, is feed a small AC voltage into the DITC. Unplug the Ignition module, and turn the ignition ON. I'll start with a 200 or 300Hz signal, and slowy bring up the volatge from zero... to about 1.5VAC. I'll be watching the tach to see if it moves.

I will also get down and inspect the GAP.
BTW... I'm metric here. Nothing but millimeters.
Back to top
View user's profile Send private message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Fri Apr 27, 2012 12:00 am    Post subject: Reply with quote

I should also add, the sensor was pulled and inspected. It appeared fine. .... but what does that really mean???
Back to top
View user's profile Send private message
ptheskil  



Joined: 03 Aug 2004
Posts: 128
Location: Essex, UK

PostPosted: Fri Apr 27, 2012 12:06 am    Post subject: Reply with quote

Matilda wrote:
I've scoped this sensor when it's plugged into the DITC


Try it without being plugged in as well in case there's something in the DITC thats destroying the signal

Matilda wrote:
I think what I'm going to do next is, is feed a small AC voltage into the DITC. Unplug the Ignition module, and turn the ignition ON. I'll start with a 200 or 300Hz signal, and slowy bring up the volatge from zero... to about 1.5VAC. I'll be watching the tach to see if it moves.


Good plan.

Matilda wrote:
BTW... I'm metric here. Nothing but millimeters.


And there was me expecting the "I'm not American I'm Canadian" speech. I must try harder.
_________________
1981 931 series2 Euro spec
Back to top
View user's profile Send private message
Matilda  



Joined: 17 Apr 2012
Posts: 30
Location: London, Ontario, CA

PostPosted: Fri Apr 27, 2012 12:48 am    Post subject: Reply with quote

ptheskil wrote:

Try it without being plugged in as well in case there's something in the DITC thats destroying the signal


I thought of that... and I think I did scope the sensor in hand with engine cranking...but I did so much that day. But I will check again.

I truly hope it's not the sensor... no one has any. NLA.

I rarely see imperial. I work for an outfit that designs military equipment, and although our major end user is the US military.... the drawings are all in mm.
Back to top
View user's profile Send private message
Rasta Monsta  



Joined: 12 Jul 2006
Posts: 11410
Location: PacNW

PostPosted: Fri Apr 27, 2012 3:13 pm    Post subject: Reply with quote

Beemer unit is on order. Stand by for specs.
_________________
Toofah King Bad
  • WeiBe (1987 924S) - 931 S3
  • Red (1987 924S) - Pro44/Spec944
Back to top
View user's profile Send private message Visit poster's website
RC  



Joined: 25 Mar 2007
Posts: 2636
Location: Australia

PostPosted: Fri Apr 27, 2012 7:11 pm    Post subject: Reply with quote

I personally have very little interest in this thread, don`t have a DTIC or even a 931. Or particularly correcting the misinformation that could be googled in a few minutes by anyone with motivation.

IMO the whole DTIC system, although hi-tech at the time, in this day and age, is the equivalent to a B&W TV. For a few hundred $$ at most, it could easily be replaced with a far more reliable & modern programmable ignition system. Not just one knob for timing but a pretty graphic interface that allows adjustment of so many other related parameters as well. But that is not to say that I don`t appreciate that some may wish to retain it for historical reasons, budget, or whatever.

Also have the greatest respect for members like McGyver who have put numerous hours into reverse engineering this dinosaur. Then others like Dan and James who go the extra mile out to the garage to measure something to help another member.

To help keep these cars on the road where they belong, I may be able to contribute a little of my knowledge and time.

Much info is available via google (patents) on the system. At least a block diagram and basic system operation.

http://www.google.com/patents/US4063539

Plenty of links to follow and some good research material for those with interest & motivation. But little time to keep going myself.

http://www.google.com/patents?id=2FswAAAAEBAJ&pg=PA3&source=gbs_selected_pages&cad=4#v=onepage&q&f=false

It is apparent that the DTIC system uses a VR pickup for the speed/ reference sensor. Thanks to ptheskil`s patient and accurate post the OP is now much better informed, without using google.

ptheskil wrote:
Matilda wrote:
A VR sensor is basically a magnet sitting in a coil. As the flywheel moves past the sensor, the magnet moves in/out of the coil and thus creating a volatge.

I suspect the magnet in mine is stuck...not moving.


Not quite. Simple VR sensors do have a magnet in them but it is static. It produces a magnetic flux which is coupled to whatever ferromagnetic material gets near it. As the teeth on the flywheel pass the sensor the magnitude of the flux changes which gives rise to the AC voltage as you correctly state. This voltage is a function of rate of change of flux - meaning that the faster the flywheel moves the higher the voltage. At crank you won't be getting much and at 6000rpm it could be in the 100s of volts.

If the continuity/resistance between the pins checks out there's not much can go wrong with it. However, if it is not positioned so that there is a very narrow gap between the teeth and the sensor face you will not generate enough voltage at crank for the DITC to make sense of it. You should try and measure the voltage across as high a resistance as you can so using the input of a DVM on VAC setting is all you need to do. As you have a scope I would expect you to find something though even if it is a very low voltage signal. Check that you are not shorting the leads from the sensor or placing a low resistance across them as that will destroy the signal.


The DC resistance of our sensor is very low indeed compared to other common units, but that simply means either thicker gauge wire (better reliability, vibration, heat tolerance) and/or less turns. It is still just turns of wire around a magnetic core, pure and simple. Less turns however will generate a proportionally lower voltage. Shielding around he signal lead always reduces EMI regardless but is more important on a lower voltage signal lead, think of microphone leads or guitar pick ups, exactly the same principle and similar voltages. Same principle as electric power generation too, coil of wire, magnet, movement. A lower inductance coil will however also pick up less EMI from other electrical items, ignition, alternator, electric motors, so sure that Bosch engineers arrived at a reasonable compromise here.

Therefore to measure any output, on a scope even, will possibly need a much more sensitive setting than 100mV/division. Used a far more sensitive VR on hand (ford) at ~900 ohms and to get reasonable deflection on the scope by simply waving a screwdriver in front of the business end, dropped down to 10mV. So keep drop down to 1mV if need be. You wont blow your employer`s scope, the front end is fairly well protected against normal overload. Also the timebase setting needs to be quite slow here, try 10ms to start with. No need to use the rotating flywheel here at all. If you can not get some deflection by moving a ferrous object in front of the sensor you have a problem.

Continuity is a poor mechanic`s term. Resistance is a quantifying measurement. Dead short is 0 ohms, fully open is infinity. 20 ohms on a 1K coil for example is within 2% tolerance but on a starter motor or battery cable is a complete fail. Now all the readings posted above on this sensor will very likely be incorrect unless a zeroing meter was used and/or adjusted, or allowance was made for the resistance of the leads. At minimum, short the leads together on the lowest setting and DEDUCT this from the measured reading of the test subject.

Quote:
On my unit, I'm measuring the following:
pins 13 to 15 0.46 ohms.
pins 13 to 14 8.47 ohms.
pins 14 to 15 8.68 ohms.


Whatever your DMM leads are, probably fairly low though, say around 0.2 ohm, well reduce figure by that much. So in the first case 13-15 will actually be 0.26 for example.

Ideola appears to have a cheaper meter with thinner leads.
Quote:
On my two spares:

Spare 1: 13-15 => continuity; resistance = 00.7
Spare 1: 14-15 => continuity; resistance = 08.9

Spare 2: 13-15 => continuity; resistance = 00.7
Spare 2: 14-15 => continuity; resistance = 08.9


Guessing up to 1/2 ohm or so. No big deal though and easily compensated for. Then all these readings will be accurate and closer to the Porsche/ Bosch specs, whatever they were. 13-15 is clearly between the shielding and signal ground whereas 14 -15 is the actual winding.

For diagnostic purposes the DC resistance is generally sufficient, no open circuit at least. It will not however accurately indicate a shorted turn. Very unlikely in a VR sensor, but more common in an overheated motor or transformer. Inductance alone will also not conclusively diagnose a shorted turn but give far better indication. There is a device specifically for this purpose, a shorted turns tester, but very rarely ever need to use it.

Particularly for comparison to BMW units or possibly just a figure we can post in this thread for reference, have a look on your DMM guys for an inductance setting. Most better meters, and lots of cheapies these days especially those with capacitance capability, will read it. Base unit of measurement is Henry. For example only, the ford sensor measured 0.384 H. Common with the SI metric system, 1000 mH (millihenry) = 1H.

Now in reality this is just basic stuff, not rocket science at all. The main factor with this sensor is the physical size. The electrical parameters can easily be compensated for with a little knowledge or trial & error & luck. The appropriate connector can easily be changed/ swapped/ hard wired. Worse case would be a basic circuit using a few $ of parts.

Was that a reflection of the twinkle in Dan`s eye?
With some community input I`m sure we can come up with an alternative available (probably budget too) substitute.
_________________
World`s quickest 924 2L slushbox

Allan @ DTA wrote:
I have no issue with superchargers, they are for guys who want to drive a car rather than talk about horsepower with their baseball cap on backwards
Back to top
View user's profile Send private message
McGyver  



Joined: 24 Feb 2009
Posts: 354
Location: Jelenia Gora - Poland

PostPosted: Sun Apr 29, 2012 3:57 pm    Post subject: Reply with quote

This is some interesting infos on down of this site about DITC and Flywheel sensor:
http://www.williamortel.com/porsche-924-turbo/info-qel.html

1,8 degree of crankshaft rotation from that sensor in 1981 ... that is something
Regarding from that site there are like two sensors ... which produce some voltage. There are 99 normal teeth and one special (which is position reference teeth) with some special ferromagnetic insert... which should produce much more voltage.
One colleague from here "michal928" made special table to test DITC, but I'm not sure he checked how flywheel sensor really works.

DITC counts impulses in some time - maybe 1sec or even less that - don't know exactly. And after passing reference teeth the counter is being reset and counts again.
There are four counters on board... two are binary and two decimal.
I'm pretty sure one part corresponds for reference time (minutes/seconds?) compared to second unit counting up to 100 teeth.

I didn't finish my work on DITC, but right now we got small baby and I really don't have any free time to make something more... I back to it later and maybe then I finis at least schematic. Which is really complicated if you want to draw it from PCB There are just 18 IC's with approx 14 legs that gives you more than 250 traces and at least 500 in whole PCB. It's really hard to draw it, and I'm still thinking how to make it good to prevent mistakes, or just how to don't skip something.




That side is complete as I remember fortunately PCB has only one layer on top of each side and can be optically checked. I had one PBC without any elements ready do be scanned and copied to PC. It's impossible to check everything with elements on board especially on elements side

For now I made complete part list, so after just few changes it's possible to made 937 DITC from any other unit. One thing I forgot was to write which traces should be cut for each unit - but that can be easily checked on photos. I sold my 937 DITC unit but made quite enough photos and measurements which I can.

Popular DITC problem consist on diodes in power 12V circuit which can be destroyed and that damage whole DITC. But that can be easily repaired with quite low cost so I think that units are almost impossible to take damage which can't be fixed with soldering iron and some free time Only problem could be that pressure sensor.
According to all documents I read DITC can work -maybe not to good... only with crank sensor connected. All other sensors are "optional"
Ok... that's enough I had to back to work now.
_________________
931 82' - 5000km after full engine rebuild

SOLD: 924 81' N/A- with turbo gearbox "dogleg" - great but mysterious car
www.mauser98k.internetdsl.pl - if someone like german rifles (English version avalible)
Back to top
View user's profile Send private message Visit poster's website
Rasta Monsta  



Joined: 12 Jul 2006
Posts: 11410
Location: PacNW

PostPosted: Mon Apr 30, 2012 3:50 pm    Post subject: Reply with quote

RC wrote:
I personally have very little interest in this thread, but that will not stop me from posting a lengthy and helpful reply.

_________________
Toofah King Bad
  • WeiBe (1987 924S) - 931 S3
  • Red (1987 924S) - Pro44/Spec944
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    924Board.org Forum Index -> 931 Tech. All times are GMT + 11 Hours
Goto page Previous  1, 2, 3, 4, 5  Next
Page 3 of 5

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group