MFI cold start saga continues

Jay Zollinger · 11-29-2005, 08:52 PM

Now I'm at a loss. Here's what I have done:

1) Confirmed that the solenoid works by connecting hot lead on solenoid to the #1 fuse in engine compartment and observing fuel flow from solenoid (once I turned on the fuel pump).

2) Cleaned solenoid ground and all other electrical connections (14-pin connectors, relay pins and sockets, etc.).

3) Confirmed that relay is not the problem by switching it for a relay known to be good (at least good enough to make the horn work).

4) Confirmed that termotime switch is not the problem by jumping the leads to the switch.

5) Confirmed continuity of wires between relay, fuse, thermotime switch, and solenoid--all of the connections that should be there appear to be there.

6) Confirmed that there is power to the #2 fuse in the engine compartment (with the ignition key switched "on").

After cleaning and testing everthing, I still don't get fuel when cranking, either with the thermotime switch connected or with it by-passed. I do get a few really quick pops of energy to the solenoid, but they are too short even to read with my multimeter and clearly not strong enough to activate the solenoid.

Grady, Warren--any ideas of where to go next? Is it possible that the terminal on the ignition switch that sends juice to the relay is malfunctioning? If so, how do I test that?

I'm inclined to wire in a separate switch, but I would prefer to make this work. Please let me know if you have any ideas. The car, by the way, is a largely stock 1973 S.

POKDUB · POKDUB's Garage

Jay,

Forgive me if this sounds obvious, but this is how I would go about it.

Can you attach a multimeter to the 12V terminal which activates the solenoid.

Crank, and then observe if it jumps up to 12V. If it does not then you have to track the feed back through the wiring.

I had problems with this solenoid. The outer casing which should be a 0V (ground) was not properly grounded, so I had to run a separate grounding strap, please check continuity between casing of solenoid and car grounding.

I also found that once the solenoid worked, it had the effect of drowning the engine with fuel, so I hooked up a separate push buttion feed to the dash. Pressed for one second before cranking, (hearing the solenoid activate), the car would start almost immediately. I know each car might be different and not as easily started as mine.

Hook up a temporary switch at least to verify before you start running an internal cable.

Rgds
Peter

Jay Zollinger · 11-30-2005, 07:30 AM

Thanks for the sound advice, Peter.

I have confirmed (with a test light attached to the #1 fuse) that the solenoid housing is well-grounded.

When I hook a multimeter up to the hot lead to the solenoid and crank the engine, this is when I get a reading of few very quick pops of energy that don't last long enough to measure on my multimeter.

I also confirmed this morning (with a test light) that the #30 relay sockey is getting power when the ignition is switched on, so I no longer think that this is the problem either.

Any additional ideas? Am I correct that solenoid should fire when the thermotime switch is closed, so that it should be working when I jump the leads and crank the engine?

Grady Clay · 09:31 AM

Jay,

You are doing all the correct things.

Check to see if relay pin 85 is getting 12 V when cranking. That pin goes directly to the starter 50 and thermo-time switch heater via pin 1 on the engine 14-pin connector. Power comes from the ignition switch 50 via pin 1 on the chassis 14-pin connector. (Note; pin 1 on the chassis 14-pin connector is wired to pin 1 on the engine 14-pin connector in the relay panel harness. These are all yellow wires.)

Check to see if the connections at thermo-time switch are correct. The yellow wire goes to the heating element (small terminal) and the red/blk wire goes to the switch (large terminal). The red/blk wire goes through pin 7 of the engine 14-pin connector and then to pin 85 on the relay.

You might check the continuity of the body of the thermo-time switch to ground.

Is there any signs of degradation of the wiring? If someone ever reversed the connections at the thermo-time switch it can have damaged the switch and the wiring. Yes, I have seen the small connector forced on the large terminal of the thermo-time switch.

Keep up the good diagnosis work. You will solve this.

Best,
Grady

Jay Zollinger · 11-30-2005, 09:38 AM

Thanks, Grady. I'll check the power to relay pin 85 when cranking and let you know what I find. I have not seen any signs of degradation in the wiring, and the thermotime swtich connectors have not been forced onto the wrong terminals. I will double check, however, to make sure that the right connector is on the right wire.

It might be that I just don't understand the thermotime switch. I have assumed that it is a simple open/closed switch that opens when the temperature hits a certain level. If one of the temrminals is a heating element and the other a switch, I assume that this is wrong.

Is the thermotime switch grounded to the chasis when installed? I have checked the switch itself for continuity between the terminals, but did not realize that it should be grounded when installed. If so, this would mean that I cannot test the system by by-passing the switch and jumping the leads. Is that correct?

Grady Clay · 11-30-2005, 10:40 AM

Jay,

The thermo-time switch is a great example of pre-war (WWII) technology. The bi-metal switch senses the temperature and disables the cold start enrichment if above a certain temperature. Additionally there is a heating element that is powered by the wire to the starter. The more you crank, the more heat to the bi-metal switch. Soon the bi-metal switch opens and prevents further cold start enrichment by interrupting the circuit from the relay to ground. The colder the case is, the sooner the thermo-time switch cools off and again allows cold start enrichment. The longer you wait between attempts to start, the switch can cool off. Thermo - Time.

Today there are 1500 lines of compiled code to accomplish the same thing.

Many thermo-time switches have never been removed from the breather casting, even at rebuilds. There is the brass housing, aluminum or copper sealing washer and the magnesium breather cover. All are subject to contact corrosion from dissimilar metals. I would take a large clip to the thermo-time switch housing and measure the continuity (resistance) to the ground post behind the fuel filter. I would also see if there is any voltage during cranking.

Have you cleaned the transmission-to-chassis ground strap?

You can test the thermo-time switch off the engine. The switch opening temperature is stamped in the housing. Use a pan of ice/water mixture to start the test. With a thermometer and wiring attached to the switch, heat the pan (continuous stirring) and note the temperature where the switch operates. Check the resistance of the heater. Apply 12 V to the heater and see when the switch opens from 0°C cold.

There are many different thermo-time switches. VW, BMW, MB, Audi and others use them. It is not unlikely that the wrong one is installed. Porsche OE thermo-time switches have the Porsche part number stamped on them. Yours should be 911.617.117.00. Additionally there is a temperature and a VDO P/N stamped in the housing. What does yours read?

Quote:

”I have checked the switch itself for continuity between the terminals, but did not realize that it should be grounded when installed. If so, this would mean that I cannot test the system by by-passing the switch and jumping the leads. Is that correct?”

If you connect the two wires at the thermo-time switch, the relay will not operate because both sides of the relay coil are now at 12 V. The best test is to ground the red/blk wire (large terminal) and confirm the relay operates when the starter is operating. Make sure this is not the yellow wire.

Best,
Grady

Jay Zollinger · 11-30-2005, 10:49 AM

Brilliant. The simplicity of the mechanical solutions used to resolve various complex problems is one of the things that I love about my car.

A lot of my testing has been done trying to bypass the switch. I will play around some more with the switch in place and/or with the large terminal grounded to see if I can make things work, and will let you know what I find.

I am pretty sure, by the way, that the switch is correct. I don't have it with me now, but it does have the correct PP#, and it is a VDO switch. I think the temp rating may be 118, but I will need to check to confirm that.

mjshira · 11-30-2005, 11:03 AM

First let me say that I feel your pain, I was there this time last year with Tyson's 2.5 in my car. What we did was wire a momentary switch and use it, rather than the factory circuit. I would simply hit the switch a few times before starting and while cranking and the problem was solved. I did have to use some starting spray a few times before this solution was in place. I was loath to do so but it worked also. One thing that helped on cold nights was to put a blanket over the engine lid when you park the car. that kept the engine from getting as cold and made it easier to fire. Or at least it seemed to.

you have the best guys addressing your question so listen to them over me.

good luck

James

Early_S_Man · 12-01-2005, 11:06 AM

Jay,

The MFI Thermo-Time Switch should be stamped 45°C, or 113°F.

The problem with using a common test lamp or Voltmeter for this type of testing is that the demanded load current isn't being tested at all, hence, marginal grounds [or connections] aren't being stressed! My suspicions point at either of the 14-pin connectors ... or the connections at the Thermo-Time Switch itself!!! A bit of Brakleen degreaser, and some 9% pickling vinegar on Q-tips can clean away years of corrosion build-up.

Jay Zollinger · 12-04-2005, 03:36 PM

It must be the switch itself. When I connect the black/red wire at the thermotime switch to a good ground, everything works as it should. I have meticulously cleaned the switch, copper washer, and mounting port on the breather cover, so there's no corrosion there that would create a bad ground at the switch itself. With the wires (correctly) attached to the switch, however, the system still doesn't work. So, unless someone here thinks I've missed something, I'm going to shell out the $$ for a new switch and try that.

(The switch is, by the way, the correct VDO 45C switch).

Jay Zollinger · 12-08-2005, 08:17 AM

I picked up a new switch, and everything works as it should. Many thanks to everyone who provided such good advice!

pwd72s · 12-08-2005, 08:39 AM

Bravo! Kudos to all who gave advice. This one I'm going to remember..."just in case".

Grady Clay · 12-08-2005, 08:47 AM

Jay,

Good for you. Everyone benefits from successful conclusions. Thanks for posting the results.

Please post all the numbers on your OE thermo-time switch. There should be a temperature, a date, a Porsche part number and a VDO part number. What is your engine S/N? I recall some undocumented changes with this switch.

Best,
Grady

Jay Zollinger · 12-08-2005, 09:09 AM

I don't have the switch here, but the temp stamp is 45, the date is 11.72, and the PPN is 911.617.117.00. My engine # is 6331141. The only change that I noticed between the old and the new switch was that the original is stamped 12V/8W and the new one (same PPN, etc., also a VDO switch) is stamped 12v/40W.

Grady Clay · 12-08-2005, 09:55 AM

Jay,

Thanks.

It will be worthwhile to document this. You clearly have a late ‘73S. Can others report their OE thermo-time switch labeling?

I think the change is to reduce the amount of time the cold start can be active. That is desirable from the standpoint of reducing oil dilution with gasoline. It is possibly at the expense of the ability to start cold.

Was this a change during production of 2.4? Was this only original with the 2.7 or 3.0 and the spare parts were consolidated?

Best,
Grady

Biff · 08-08-2006, 05:58 PM

I'm in the process of trouble shooting the same problem. My car is a '72 built in 7/71. The markings on my thermo-time are: 12V3W; VDO36/3/4; 911 617 117 00; 4 70; +45`C. My thermo-time has a paper/hard compressed cardboard washer not copper or aluminum like Grady stated. Could that give me a grounding issue and not let the thermo-time switch work... and then not let the cold start injector work??

Biff · 08-09-2006, 06:53 PM

Well it had to be the cardboard/paper washer was not letting the thermo-time switch ground and not allowing it to complete the circuit. Replaced it and the entire cold start system works flawlessly!!! I would have NEVER guessed that one if I hadn't read what Grady mentioned about aluminum or copper washers on thermo-time switch and the fact that thermo-time switches are never touched or cleaned. I don't think many would have found that one...many thanks to you mfi experts!! WOW!
Biff