Introducing my timing light - injection timing- digital RIV method- bench test
 

The FSM procedure 07.1-108, Checking injection timing (begin of delivery) with digital tester (RIV method) has a diagram which shows a setup of 3 pieces of hardware for checking injection timing.

http://mercedes.thatchermathias.com/w123CD2/Program/Engine/617/07_1-108.pdf

The 3 pieces of hardware are:

617 589 10 2100 RI- Transmitter
617 589 09 2100 RI- Adapter (for available digital tester)
Bosch MOT 001.03 Digital tester

I have built my own system which provides the same functions as the above. Just did a bench test and took a short video.

On the left is a drill spinning a disc with a screw to simulate the tang behind the injection pump timing plug.

Clamped in the vise is the pickup (RI-transmitter)
The pickup is plugged into the RI- Adapter (The aluminum box) which outputs to an inductive timing light.
The digital tester is just a conventional inductive timing light.
http://i31.photobucket.com/albums/c3...303_153627.jpg

CLICK TO PLAY VIDEO!
The portable drill goes up to 1000 RPM. As you can see, tracking is spot on at all speeds! The strobe cutout you see is due to the drill not hard mounted relative to the pickup so the gap moved a bit as I operated the drill. I will be testing this in the car over the next few days and will shoot another video.
http://i31.photobucket.com/albums/c3...303_154136.jpg

:dancefool

Nice work! That is awesome :). Should help with IP work.

Thanks!

Very cool!

Hmm
 

Following this experiment.

.

How much did the transmitter and adapter cost? Where did you source them?

That's cool! As a variation on the idea, I'd like to see a test injector rigged up and the timing light pointed at the spray to see the degree difference between the timing signal and the actual injection event.

Impressive! Outstanding results so far.


This is an excellent idea!

I made them myself.

Interesting test. How do you propose I set this up? Cranking or idling?

We are jumping the gun here. Although I am confident what I built will work on the engine, I have not actually installed it on the engine yet.

Understand that you are still in development. I'm just brainstorming possible applications.

I envision a spare injection hard line, re-bent to run to an injector aimed into a large glass jar. You could connect this to the # one delivery valve and just run the engine at idle on 3/4, 4/5 or 5/6 cylinders.

My timing light has an advance adjustment dial so I also imagine that while the engine was running with your trigger mechanism attached, you could adjust the advance dial and shoot the injector spray until you could see the very beginning of the injection event.

This exercise would tell you the exact offset between the position of the trigger lug in the pump and the actual injection event and would take into account all of the variables - actual port closure, hard line flex, pop pressure, etc...

Then, you could shoot the crank indicator with the freshly calibrated timing light and adjust the pump until the desired indication is reached.

Good work. How do you trigger the timing light - with a low voltage signal or with a high voltage? For the later, one could use a GM HEI module to drive a coil, and clamp over the HV spark lead as the timing light is normally used. A VR type crank sensor can trigger an HEI module. A round-about way, but with all off-the-shelf parts (have in my garage).

Reading between the lines of the M-B instructions, it appears their processing box times the delay between the TDC pulse and the RI pulse, and converts to crank angles. It seems you could just flash the timing light at the crank marks to verify it flashes at 15 deg ATDC. It appears that the RI pulse does not align with "start of delivery". Perhaps it marks TDC of the #1 IP piston.

I borrowed a Ford timing light setup for my 85 F-250 with International 6.9 L diesel with a Bosch injection system. It used some kind of inductive clamp on the fuel line to operate the light. Anyone tried it on a Benz?

Yeah, I have one made by Ferret but the clamp is too big for the line - the 606 has, I believe, 4.5mm lines and the clamp is designed for 6.25mm or some such. I put a collar around the line to try and make it work but the results are not encouraging.

Your test would reveal some useful, non existent data on our engines. I do not have a timing light with advance currently and cannot do the test. Maybe when I have a working unit, I can let you borrow it and you can do the test?

Bad news:

I suffered a setback today. While installing the sensor into the IP timing plug, I damaged the wiring inside the sensor. It is not salvageable and is now junk since it is potted in epoxy. I have to come up with a better sensor design. What worked so well on the bench is no longer working on the bench. Damn, a lot of work went into that sensor! I have more parts on order which should arrive in a few weeks.

Ideally you need to hook the trigger sensor to the timing light and point that bugger at the graduated marks on the crank - as simple as that isn't it?

Shame that didn't work out.

KarTek's test goes beyond that. Yeah, the damn wire within the sensor broke before I was able to turn power on. What a waste. I'll be back.

Stretch, on the last page of FSM 07.1- 010 operation injection pump with governor, can you translate the labels in the timing diagram? Some of them are in German I think.

http://mercedes.thatchermathias.com/w123CD2/Program/Engine/617/07_1-010.pdf

If I'm understanding your project correctly, you're attempting to emulate a gas engine timing system with a diesel. You've been able to use a VR sensor to trigger a Xenon timing light on the desk. The plan is to use the IP as a trigger wheel and drive the timing light to read the timing on the front pulley. The tang on the IP you're attempting to read should read 15 degrees ADC if the timing is correct. Is that correct?

In thinking about the system I wonder if it is possible to use either a piezo pressure sensor or maybe even a piezo knock sensor as a trigger event for the timing light. The commercial systems use some kind of a clamp on the injector line that read the expansion of the injector line as pressure to trigger the timing function... at least that's the way I understand their operation. That would take a pretty sensitive sensor to read the tiny bit of movement of the line wall.

I'm wondering if it might be possible to create a more responsive and cheaper system that would use a high pressure piezo sensor or maybe even a knock sensor. What I'm thinking about is two sections of stock injector lines (red) connected at a block of aluminum, brass or steel to the #1 injector outlet of the IP. The green connectors in the below image are modified fittings from the IP. I've made a bunch of these fittings for use in a pop tester - metric injector fittings on one end and 1/4NPT on the other to make fabricating a connector block extremely simple. Reading the signal should be pretty easy, although I think it would be a rising edge rather than a falling edge you'd be looking for. If you can use a knock sensor they're dirt cheap - a pressure sensor, maybe not so much.

http://www.peachparts.com/shopforum/...1&d=1394243156

I may be completely off base with this idea, but who knows, maybe it would trigger some other options to accomplish the goal.

If you're using a VR sensor, you can use almost anything that would fit. RPM and speed sensors off the MB, cam position sensors, wheel speed sensors, and so forth.

OT means zero degrees TDC
v.OT means before TDC
n.OT means after TDC

Was that the bit you wanted?

This approach has essentially been considered here =>

http://www.peachparts.com/shopforum/diesel-discussion/349705-injector-line-pulse-frequency-help-needed.html

The difference however is to not arse about putting in pressure transducers (you'd have to bleed) but to measure the elastic deformation of the injector line (clip on clip off)

Instead of a two-part hard line, a small fitting for the piezo could be machined that either screws onto the injector, or the IP fitting; and the hard line screws onto the fitting. You would just leave it always in place.

Yes, thank you.

On the top chart, below the left most vertical scale line @ 24 deg v. OT, Is the = sign in "OT-I = RI = FB - P" meant to be used as a formula?

What does "FB" and "P" stand for? Is the "-" a dash?

I'll fire up (bloody) windows for you tomorrow and see if the original language FSM sheds some light on it all (pun intended)

Sorry for the late reply. I've been busy. There are many ways to skin a cat and I see no reason why your method will not work. It may produce different results and you just have to characterize it with testing and measurements. I say go for it!

Bump

That formula bit if it is a formula doesn't exist on my copy of the German language FSM.

I think it should be ignored.

Good to know thanks! Those Germans trying to confuse us?

I don't know what it is - may be they were trying to figure something out.

you know...
 

you've got something really interesting going on when Roy is following your thread! Yeah, this one is most interesting!