Tach amp repair attempt
 

For a challenge, I'm trying to repair a bad tach amp for a friend.

I picked away at the white silicone potting compound and found a badly burned out compoent that I cannot identify. Is there shematic of the tach amp somewhere? How about a wiring diagram of it in the car? I was not able to find a new one on Fastlane. Is it called something else? How much is a new one?

Here are a few poor quality pics I took while taking the tach amp apart:

http://i31.photobucket.com/albums/c3...052607_002.jpg
http://i31.photobucket.com/albums/c3...052607_003.jpg
http://i31.photobucket.com/albums/c3...052607_004.jpg
http://i31.photobucket.com/albums/c3...052607_005.jpg

This is the 83 300D part:

http://catalog.worldpac.com/mercedesshop/sophio/wizard.jsp?partner=mercedesshop&clientid=catalog.mercedesshop&baseurl=http://catalog.peachparts.com/&cookieid=1CQ0J3JZ423G0KC1AE&year=1983&make=MB&model=300-DT-001&category=P&part=RPM+Sensor

enough to at least justify picking it apart :D .

I sent a well picked apart one off to another member several months ago who was going to dissect further. No word as of yet, but I will say with the amount of silicone you've removed there's still a bit to go (photos are of pretty poor quality too so it's really hard to see). Some components are marked, some of them are not (the green ones if I remember correctly).

on any non-functioning tach the cigarette butt tricked should always be attempted first.

$$$ Ouch! Why is it called a RPM sensor? Technicaly, the Tach amp is not the RPM sensor. The magnetic pickup on the engine is the RPM sensor.

Anyone know whether the tach in the instrument cluster have any electronic components?

Sorry for the poor quality pics from the phone cam. The phone is always on me, unlike my digi cam. Are poor pics better than no pics?

The tach gauge I believe works very similar to the fuel gauge. It receives a signal and places the needle based on the signal. Fuel gauge works on varying resistance. Not sure what type of signal the tach amp sends but the gauge does no 'processing' of it.

bad pics better than no pics? all depends on the context.

Would you consider getting yours back (if possible) and send it to me for repair? If yours is intact, along with the one I'm working on, and I can draw a schematic from them, and there are no proprietary components like an i.c. there is a good chance these things can be repaired or a new circuit built to replace it.

member PerkHouse has it and I've not heard from him in a while. Your best bet is to get with him and others from this thread to figure out the best approach.

obviously this is not the first time the subject has come up. ;)

I'm sure it hasn't been the first time this subject has come up because this is my problem too!:D Mine works intermittently so I think I might be able to repair it. Now I've been told you could fill the top with pennies on top of the silicon and it will work for awhile. Something is making a count of the electromagnetic pulses and converting that to a consistent revolution monitor that can expand and contract the engine speed against the electromagnetic pulse. My guess would be a capacitor and a toroid with a voltage division network coupled to a threshold diode. I haven't taken mine apart yet, but when I do I'll post back and let everyone know what I found. I've got quite an extensive work bench with test equipment for restoring the circuitry on Antique tube type radios and this knowledge comes in handy!:D there is a schematic of the electronic ignition uniot on Chrysler vehicles from the 1970's and early 1980's in almost every auto repair manual and you might be able to work that backwards since values are given for the components.

Bad news! There is a proprietary IC chip in there and I can't identify it! It's not in my Phillips semiconductor catalog and I cannot reference it to any component link.:eek: I think I now know why these things are so high priced! I know that there must be a way to find or make another type of broadcast sensor and I'll keep working on it and post back:book:

I have the silicone potting mostly picked out. Here's a pic

http://i31.photobucket.com/albums/c3...052707_002.jpg

The i.c. is a LM1815N on mine and AFAIK is not proprietary. Yours must be different.

Here's the internal workings of the LM1815N, see picture.
Made by National Semiconductor, costs about $7 in an electronic supply store.

No it's the same as yours, but mine had no visible lettering on it so I assumed it must be proprietary and it isn't! I found the circuitry to this system and I'll post it here shortly.:D

I found the schematic for the RPM sensor! I'll have to start a new post on this topic because I cannot upload a bmp file to the post reply section of this forum. for those interested look in the headliner VIOLA!:D

VIOLA! RPM sensor schematic found!
 

Now can anyone tell me how to upload a file bigger than 65kb?

What type of file? Images, use http://imageshack.us , others, I can host it on my server if you want

Here!
 

it is. I cannot get this forum to post a file picture of this size so all I guess I can do is send it to anyone who wants one,PITA!

upload it to http://imageshack.us and then post the url here

OK, how do I send it to you?

email it to [email removed]

OK, I could not post it to imageshack! this is basically a scanner copy of the printed image and I can't resize it to fit here, BIGGER PITA! 300SD81, I sent it to you and if you can host it on your server I'm sure the members would be greatful. I'm glad to have found this so I can share it here. I think everyone has had this problem of their RPM sensor conking out!:mad:

I'm happy to host it as well: cwyatt at chuckwyatt.com

OK! I'll send it to you too! :D

here it is.. http://img508.imageshack.us/img508/7...msensoreq9.gif

I will take a copy too if anyone wants to email it to vickerymarine@yahoo.com....or, if someone can figure out how to post it here, that will work too.
Thanks!
Jeff

Disregard, it just popped up ...thanks to 300SD81 !!!

Its not truely a schematic, the logic gates have to be implemented somehow, which is not shown.

I too would like to pitch in and see if it can be replicated. I believe it is simply a pulse amplifier, the tiny signals from the sender need to be raised in amplitude, then the instrument (tach on the dash) integrates the pulses into an analog for the galvanometer. ?? what I meant is there are two steps in the chain going from the magnetic revolving thingie at the harmonic balancer then those signals are converted from variable frequency to a variable analog. Called a F-to-V converter in the electronics trade
Whatever the heck it is, that VDO module is a poor design for some reason, I always suspected a diode or something cracks due to heat or vibration or a pin in the connector fails to make good connection in the receptacle, thats why putting cig butts or chewing gum or whatnot in the top seems to fix a few of the dead modules but in my experience this is only a temporary solution as they go bad again sometime later.

www, you don't need a gate, the pulses need to be "on" consistently. I bet the pin that gates it is grounded in the car. Someone had an idea that switching the tach off for some reason? I'd use a CMOS buffer and design out the gate function.

lol, I'm not that great of an electronics guy (yet), only a junior in highschool, who designs a few circuits when I need them.. never even worked with logic stuff before.. :D

..

There is a transistor in the circuit that I will try to diagrahm out and put it here too. I know this is not a replica of the original part installed in your Benz's, but the basic concept is there. Now that I have this, I can finish the rest of the schematic, the one problem I see with this setup is the capacitor in the circuit is only 100MF at 16Volt. The vulnerability here should be obvious! This is in the original part and not on this map yet. Give me time! I had a hard time finding this much so far and you can bet that MB did not want to give their circuit!:D:book:

That is not the schematic for the RPM sensor!
 

I hate to tell you this but what you found is not the schematic for the RPM sensor! Tomas already posted the same diagram in a post above, which is from National Semiconductors data sheet for the LM1815.

Here is the pinout for the LM1815

http://www.national.com/images/pf/LM1815/00789301.pdf

I suggest you work from this pinout and not the one you found.

I hate to tell you this but this is not the schematic for the RPM sensor! Some people just like to jump the gun. :rolleyes:

Maybe $7 in Sweden? Over here it's probably under $2.:)

This is not the schematic for the Mercedes Benz part and i stated that, but this IC is a analog amplifier and this schematic is for a typical setup for this type of circuit. I would welcome help drawing out the circuit from those who have it apart as I'm working on this same part as well and drawing as I go. However some components cannot be identified, but so far those are capacitors and I'm sure that this mathematical problem can be worked out. I'm still working on it and am about 1/2 way through. didn't mean to mislead, just got excited.:book:

funola,

Why not get one from a wrecking yard or a used one from a supplier of used MB parts?

P E H

See the first post in this thread. ;)

Funola,


OK U like a challenge.

P E H

Well, this part has a high failure rate and Funola got me thinking about how to figure out the schematic and build a better one. I've been at this all day but I'm getting close to being done!

The part that burned to a crisp on mine looks like it is a resistor and it is at the +12V power input. The solder connections at the resistor also failed. The high failure rate of the RPM sensor (I like Tach amp better), based on disecting it and looking at the way it is built is IMO due to heat and mechanical stress. The engine compartment gets pretty hot, with the insulative property of the silicone potting compound, the components inside is overstressed. If the RPM sensor is mounted lower instead of where it is it would be in a cooler environment and may have a higher MTBF. 2 of the pins on mine were pushed in half way and is due to the design of the connector. Basically the pins are only held in place by the silcone sealant.

I have not drawn out the schematic yet. My eyes are kinda tired right now. Maybe tomorrow. Knightrider, if you made a schematic, take a pic and post it and we will compare.

I suggest removing this diagram since it is not representative of the Tach amp aka RPM sensor.

I'm still working on it, but my eyes are kinda tired too. I thought about taking down the post of the schematic I found, but 300SD81 is the one who is hosting it and there are some using it to figure this all out and I cannot remove someone elses post, it has been useful. we'll get together tomorrow! I'm really tired. I've been at this all day!:o:dizzy2::book:

guys -

an LM1815 IC is used in automotive applications mainly to turn a VR signal into a square wave.

I am pretty sure, the pickup & sensor on the engine are variable reluctance (magnetic), but the tach is looking for a 12v square wave (typically found on an ignition coil in a gasser - it's triggered by negative pulses) so the "tach amplifier" is most likely to convert the sine wave into a square wave (so they didn't have to design a new tachometer perhaps?)

Someone should put a shop scope on both lines (coming from VR sensor, and going to dash tachometer) and I bet there's a sine wave on the engine sensor, and a square waveform on the dash tach signal line.

I sold my '83 300D turbo last year, moved to southern CA, and I am shopping for a replacement for a DD. So i've been back on the forums doing some reading... i do aftermarket engine management for a living and I am familiar with LM1815 circuits, and other VR signal converters.

If I am right, you should be able to wire up a cheap GEneral Motors "HEI Ignition Module" between the sensor and dash tach, and get the tach working.

-scott

OK, I need tach help, too. I do not believe my problem in the tach amp, as I have several, and there is no difference with any of them. What are the odds that they are all bad?

I'd like to start with the magnetic pickup and check that. Does anyone have a systematic diagnostic procedure that wll let me check the mag sensor, and its harness? How about a method to check the tach indicator in the dash? I hate to "fix" components that are good.

My symptons are easy--the tach is dead. It never even flickers. Needle stays on the "0"mark regardless.

It is unlikely that your Tach amps are all bad. Start by taking measurements at the Tach amp socket and report your readings

1. key off. Resistance across pins 8&9. Mine is 80 ohms. I believe is the mag pickup coil resistance.

2. key on (glow light on). across pins 2&6 (2 is gnd). 12V on mine. This I believe is power to the TA

3 key on (glow light on). across pins 2&3 (2 is gnd) . 6V on mine.
This I believe is coming from the Tach in the cluster.

These readings are from a working tach in my 83 300DT

I successfully repaired the Tach amp!!!
 

Not being able to read the i.d.on the burned component, and since it is in series with +12V, I replaced it with a blocking diode, re-soldered the failed solder joints at the burned out component, I measured all the pins to make sure there are no shorts, plugged it into the socket, started the engine, revved it and the Tach works!!! Yeak baby! Here's a pic of it during testing. When I put it back together, I am not gonna pot it with silicone since it will run cooler without it.

http://i31.photobucket.com/albums/c3...052807_001.jpg

Just for the hell of it I dug out my old Tektronix scope and looked at the output of the Tach amp

Here's a pic but I don't think you can see the 2 pulses. It is a negative going pulse approx 1 ms wide at &pprox 6V p-p. Amplitude is contant with rpm change. At idle, the period measured was 82 milli-sec, which works out to around 731 rpm. (1/0.084 *60). The scope is probably not well calibrated but I'm gonna raise
the idle just a bit since it sounds low. I took a short video clip also. I'll see if I can upload it Photobucket and post a link later

I don't see it.

I'd be very interested in both, the signal at the sensor and the output to the tach.

Nice job on the repair, finally getting somewhere on this long running project.

Still interested in seeing a full on remap/schematic. I suspect your burned up component may be an isolated incident whereas many of the failures are due to a loose component that is temporarily fixed by the now infamous cigarette butt trick.

You electrical geeks are blowing my mind with this tech talk (obvious capacitor problem? blocking diode! WTF?!). Who's going to explain the concept of the square root of -1 next?

It's kinda fun for us tech geeks and I'm one of the worst! I still live in the old vacuum tube era and all my equipment and even my stereo is full of tubes! 7868 linear power pentodes in class AB1 with 9.41% inverse feedback error correction to be exact! 50 Watts per channel that can be heard down the end of the block and whe it's cold outside it will heat the house!:D I'm still working on the schematic end of this project and i hope to have it out on a post soon as I found the typical operation analog service schematic for the LM1815N semiconductor and it's operating characteristics. I'm hoping to come up with a more rugged and durable pulse controlled modulation inverter assembly.:book: