CSV not lighting up on cranking...need other parts ID

[teahead]

So, I guess there's a relay and a thermo time switch?

Which one is the relay for the CSV here?

Also, is this thing that's on the LEFT side of the head, towards the firewall, the thermo time switch?


Need to test both of these to see why the CSV is not lighting up when cranking.

[ja17]

Yes on the thermo-time switch. The relay is one near the brake booster.

[teahead]

Okay, can the relay be tested on a bench like any other relay?

are the prongs normal, 87, 86, 30, etc?

[ja17]

Yes, you can bench test these. #85 is ground (-) to coil in relay, #86 is positive feed for coil in relay,  #30 is positive from fuse box, #87 is outgoing positive to starting device.  To test relay hook ground to #85 and positive to #86. relay should activate. If contacts in the relay are clean you should have continuity between #30 and #87. You can open the relay and clean the contacts if needed.

When relay is installed, it is activated by negative (-) feed to relay coil #85 from tts. The positive feed (+) to the relay coil is supplied when the ignition key is turned to activate the starter.

If the relay is good, get back with us, for the tts testing procedure.

[FresnoBob]

Hi Rob,

I recently installed a momentary switch under my dash to activate my CSV.  The thermo / time switch & relay should activate both the CSV and the enrichment solenoid at the rear of the injection pump.  The Starting Aids Tour has excellent advice about how these all work.  You may also have a round 1 second relay that activates the starting aids when the engine is warm enough that the thermo / time switch isn't engaged. 

Good luck, I've been doing this for nearly a year and don't yet have the car starting as I would like. 

[ja17]

Looks like Rob has a late 280SL, it will not have a round one second relay. In fact, the very latest version (version V) had no starting solenoid on the injection pump. I suspect Rob has version IV.   

[Shvegel]

In the case of your car a 4 pin relay is a 4 pin relay so you can swap with the other one you posted if the cold start one is bad.  As for diagnosing the cold start valve I gleaned this from one of my earlier posts.  If your thermal time switch is bad the original is ridiculously expensive However Bosch  makes one for the air cooled 911's that is nearly the same spec and far cheaper.  You unscrew the studs on the Porsche part and screw in your screws.

https://www.pelicanparts.com/More_Info/91161711700.htm?pn=911-617-117-00-M100

What you have is essentially 2 circuits here with each running between the wiring terminal and ground.  The terminal are stamped "G" and "W" near the screws.  Remove the wire from the "G" terminal (small screw) and measure resistance between the "G" terminal and the brass body and you should have about 50 ohms of resistance at all times warm or cold.  Now for the tricky part.  With the engine at room temperature {20 degrees C or 68 degrees F) and the wires connected connect a volt meter set to volts DC between the "W" terminal (large screw) and the brass body(ground).  Remove the coil wire from the ignition coil and have someone turn the key to start  and watch the meter.  You should see battery voltage with the key on before cranking then less than half of one volt for 8 or 9 seconds and then the meter should jump up to 11 volts or so..  If you see more than half a volt it means the contacts in the thermal switch are bad and if you don't see a jump in the 8 to 9 second range it means the contacts are not opening due to there either being no power applied to the "G" terminal during cranking or there being no power applied to the cold start valve.

The last test I outlined is what is called a voltage drop test.  It is a little more accurate than just measuring the resistance between the "W" terminal and the housing because it is testing the contacts in the switch with a load on the contacts.

[teahead]

Relay is good.

However, weird results from your thermo time switch test.

Resistance between the "G" terminal and the brass body is about 34 ohms.  G-wire is disconnected.


While cranking, I get an immediate 10v throughout the whole cranking cycle.  Engine is cold, but it's about 75 Degrees in the garage.

[Shvegel]

I don't think the resistance is the factor as the heating element ("G" to ground) opens the circuit between "W" and ground.  If your coolant was under about 95 degrees then your Thermo Time Switch is bad.  You can test the rest of the system by grounding the "W" wire (again large screw), remove the small hex screw for the face of the cold start injector and crank the engine with the coil wire off.  Fuel should leak out of the hole where the hex screw was when the car is cranked regardless of temp.

As a final check make sure you have battery voltage at the “G” wire(wire disconnected) when cranking.  I am positive you have a bad thermal time switch but these 2 tests will check the rest of the system as well.

[teahead]

I replaced the TTS.  Still no voltage at the CSV when cranking.

 

I'll test again if I get voltage at "G" while cranking.  I.e. 86 should be hot while cranking.

When it's cold, wouldn't "W" be grounded?

[ja17]

Power to activate the CSV comes from fused #6 then through the relay when activated, then directly to the CSV. Pop the cover off the CSV relay and clean the contacts, next turn the ignition on and depress the relay manually. You should hear the CSV activate. Next try using the ignition (starter actiivated position) to automatically do the same.

[teahead]

Okay, I think I found the problem.

The CSV connection on top.  The terminal there goes to GROUND!  I.e. the wire doesn't go to ground, but the terminal the wire attaches to goes to ground.



So guess I have to take it apart and see why the terminal on top is touching the metal body.

[Shvegel]

Remove the screw and check the resistance again.  Just the coil is available from Mercedes and Bosch.  Try Googling the part number on the side.

[teahead]

Remove the screw and check the resistance again.  Just the coil is available from Mercedes and Bosch.  Try Googling the part number on the side.

Yup.  Wire and screw removed.  Terminal definitely goes to ground (ohm meter sounded the alarm).

Guess it's not a serviceable part.  I'll order just that part.

[ja17]

Try touching it with a hot wire just to make sure.

[teahead]

Try touching it with a hot wire just to make sure.

I can hear it click. Does that mean it's working as it should?

One thing I noticed is that with the wire disconnected, measuring voltage at the wire while cranking shows like 10+ volts for like a second or two

[ja17]

Yes, that means the solenoid is functioning. Sounds like your issue is before the solenoid. Follow my post on removing the cover of the relay.

[teahead]

Yes, that means the solenoid is functioning. Sounds like your issue is before the solenoid. Follow my post on removing the cover of the relay.

Will do, but I did bench test the relay earlier.  Maybe not getting full juice from 86 when installed?  I did test that with a test light on the relay plug and it lit up while cranking.

Normal for that CSV terminal to be grounded before applying voltage to it?

[ja17]

The "W" terminal (larger screw) should become grounded when the coolant temperature is 95 F or below. When grounded the relay for the CSV should activate when you engage the starter. You can make the system think it is cold for testing (below 95 F), by running a temporary ground wire to the "W" terminal.

[teahead]

THanks Joe!

I cleaned up the insides of the relay and now I got power to the CSV!

Interesting when I bench tested the relay, seemed fine, but when I had the cover off, but connected the car and started it, it SOMETIMES lit up the CSV, but most of the time not.

So after spraying a bunch of electrical cleaner on there, seems now much more consistent.


QUESTION:  how long cranking is it typical before the engine fires?  Still seems like on the 1st try, it takes about 5-6s.   It fires, coughs, and dies.  Then it starts up right away on the 2nd try.

[Mike Hughes]

Mine is a 230 and yours is a 280, but I find that if I turn on the ignition and let the fuel pump circulate fuel and pressurize the lines while I am putting on my seatbelts and switching glasses it will usually fire up on the first turn of the key.

[teahead]

Mine is a 230 and yours is a 280, but I find that if I turn on the ignition and let the fuel pump circulate fuel and pressurize the lines while I am putting on my seatbelts and switching glasses it will usually fire up on the first turn of the key.

That makes sense.  Will try that.

Later Benzes, they had a "accumulator" that held pressure for cold starts. 

[ja17]

Next check the enrichment solenoid on the IP if your car has one. It should activate when the starter is activated.

[teahead]

That really helps!  Turned on ignition all the way before crank for about 5s.

Engine then fires in about 2s of cranking.  Way way better than before!

Thanks fellas!  Been extremely helpful!

[thundata]

I have a 65 230sl and have noticed the same behavior. If I let the pump run for a several seconds it fires up on the first crank. Is this the expected behavior? Should the CSV allow the car to start quicker without letting the fuel pump run for an extended period? Is your CSV functioning properly? Hoping for insight before I go about diagnosing and trying to fix something that isn't broken.