DME Relay: The final word

[ebradway]

Today I swapped out the DME Relay in my '86 Targa. Because of the ongoing debate, I decided to investigate the device a little more thoroughly. I have included three pictures: one of the coils inside the DME Relay, one of the manufacture date of the device (note: my old one has a manufacture date of 4-29-1996, obviously NOT the original), one of the bottom of the circuit board with a few things highlighted.

I conducted this exploration under the auspices of my friend David, who, in his own words "I am not an electronics technician, I merely played one at NASA." We reviewed the schematics and the potential issues with my old DME Relay (to his experienced eye). The DME Relay receives signals from the DME and the Alarm System. Based on those signals the coils charge and complete the connections providing power to the fuel pump and the O2 sensor.

David pointed out that relays usually fail because of faults in the soldering or corroded connectors. The actual electronics in the relay aren't prone to failure. You can check a 'dead' DME relay by putting an ohm meter across the connectors I labeled in the picture. The readings of 73.2 ohms across one coil and 82 ohms across the other should be about correct. If your coils ohm out like this, then you can repair your relay.

To repair your relay you should check for corrosion on the external connectors. Mine are nice and shiny so we had to look at the bottom of the circuit board. David pointed out that the solder connection for one of the coils has been heated excessively - hence the striation in the solder - and cracked probably causing intermittent faults. To correct this, we simply touched a hot soldering iron to each solder joint long enough to melt the old solder and reseat the component leg. If you simply have corrosion on the connectors, use an emory board to polish the leads.

There is a potential for poor performance from a bad DME. Intermittent or bad solder connections can add reistance to the circuit. The relay is desinged to provide 12 volts of power to to the fuel pump and the O2 sensor. Since the O2 sensor is simply a variable resistor altering the voltage returned to the DME, increased resistance in the DME relay would result in incorrect O2 sensor readings and ultimately improper fuel mixtures.

Hopefully this will put to bed the entire DME Relay question. Yes, a relay should work or not, but there is the potential of generating incorrect voltages - and varied performance. However,
replacing the DME Relay completely is not necessary as 'weakened' relay can be easily refurbished. David and I discussed creating a solid-state transistor based relay but decided the cooling issues outweigh the reliabile gains.

[Joeaksa]

Eric,

You forgot the most important part! Did you feel any difference after you worked the relay over and put it back in the car? Would be interested to see if you noticed any change.

BTW, my car used to be the same color as yours. Do not see many that color. Too warm for the South...

JA

[ebradway]

I didn't notice anything appreciable. I had some intermittent issues with the engine dying- which was why I decided to dig into it. Someday I'll have to try the two head-to-head. Maybe sometime with some of the Atlanta Pelicanheads...

[964Cab]

DME revisited:

My '91 C4 Cab has an engine dying problem, can happen at dile or while on the hiway.
If at idle the engine just dies (can alwasy restart at the key) but on the hiway it shuts off and then restartes itself with a sudden jerk motion.
When the problem happens no lights are affected.

I suspect the DME relay system. I am on my 3rd new DME and still the problem exist.

The DMe relay has two coils as noted before.

One coil seems to feed the fuel pump but another coil seems to control the former coil and it's the latter coil that I suspect is "telling" the former coil to shut down. That latter coil goes to the anti-theft system, battery volatge, and the DME. So anything anything could be happening that's causing this engine to shut off.

How can I bypass the DME relay so my fuel pump always works when I start the car regardless of the status of the anti theft situation or various other false positive alarms ?

many thanks,

Patrick

[mwbaum]

You compared your new relay to one that was almost new itself! When I swapped my 16 year old original relay with a new one the car ran smoother and idled better. I think maybe your comparisons should be made against a 16 year old relay to be of value imo.

[carreradpt]

I have been following this debate for a time myself. I recently purchased a spare DME as I didn't want to be caught short if my present one failed me. Anyway, after it arrived (from pelican of course) I just had to install it and see if there was a difference. Without a doubt, there was a difference in performance. Smoother idle and acceleration. Mine was original, mid 84 date on it. I was skeptical but am now convinced. And I have a spare in the trunk.

[Wayne 962]

Anyone done the "Pepsi" challenge? We sell a lot of parts. I love it when people buy a performance muffer (that weighs more than stock), and then 'claim' by the seat of their pants that they are getting 15-20 more HP (popular claim with the 914 Bursch mufflers).

I know these mufflers, and I have dyno tests on them. They don't increase your HP, yet the people who buy them will swear that they do. This is what I call the placebo affect, applied to cars.

I need someone to have a friend or their wife mix up the old relay, and the new relay, and drive the car with both, without knowing which is which. Then, and only then will I firmly believe that the DME relay can make a difference in HP/idle/driveability performance.

I'm not saying that it's not possible, I'm just saying that the placebo affect is too powerful to generate any useful results by saying "oh yes, it runs much better now..."

-Wayne

[964Cab]

Hi Wayne,

Speaking of DME relay driving impressions how can I bypass the damn DME relay so my fuel stays on when the ignition is ON ? For some reason the DME on my 964 '90 likes to "hiccup" and drop the fuel pump hot volatge off. That requires a restart at the key (at idle) or if driving the car restarts itself like a "push-start" jerk effect.

Thanks,

Patrick

[Techspert]

Guys,

I have an '85 Targa. I too have had the engine dying out of nowhere issue. My car would unfortunately not start for several hours or several days when this happened. I have replaced the DME relay and also clean the several contacts in the engine compartment. My mechanic pointed them out and said they control air fuel mixtures or something to that effect. Since I recently bought my first home - w/garage , my Porsche has not seen much action. I will assume the DME has everything to do with my car dying on me.

One thing I did was putting a balloon w/silica gel packets around the DME relay switch. The reason behind this is because Targa top leaked and slight amounts of water would be on the floor near the DME and processor box area under the driver's seat. The balloon and silica is to protect the DME and keep it dry. I also put a couple of silica packet in the Processor box. I noticed that after period of heavy rain, my interior would have condensation. This was when the dying would most occur.

Am I in the right direction with my assumptions? Thanx...

[mm86911]

Techspert - you may also want to stash some dessicant in a sock behind the dash gauges - should reduce chances of them fogging up during humid conditions.

[mwbaum]

Smatta Wayne?? Doncha wanna sell more relays already??

Well having lots of experience testing/replacing relays on F-4E/G's to 8-lamp systems on school buses.

1. You need a good ground for the coil

2. You need good volt's & amp's (no corroded wires connectors etc) for both signal and switched power

3. Measure the relay with an Ohm meter across the switched contacts. You will have to look at the pin out to see if the open and closed positions are used. You should measure less than 1 or 2 Ohms resistance.

4. A relay can be thought of as a set of points (sorta realy slow ones) both have contact surfaces that can pit corrode etc which increase resistance drop volatage and up the Amps.

5. If you have a problem and a new one corrects it..... take a hammer and smoosh the old one. And yes new ones can be bad from the factory.

6. If you suspect wiring check it point to point for resistance with components on both ends disconected.

So it could be anything from the signal source (keyed power, ign switch etc) wiring etc. Also check connectors for corrosion on the pin side AND where the wires enter the connector.
Also when checking the connectors don't use a probe larger than the female connector.
Well thats more than what I thought I was going to say.
Hope it helps
Chris

[964Cab]

I too have water leaking into my 964 and the 993 Cab through the door weather striping
and getting to close for comfort to the $$$$ DME brain under the driver seat. Here is how I cured the
water leak. I bought 3/4" foam weatherstriping from Frost King (sold at every Home Dept, etc) It's
3/4" wide foam striping and 3/16th thick and comes in 17 feet rolls. Part number Frost King V4498. Take it and
use weather strip glue to secure all along the edge of the door frame (not the body frame where the stock factory rubber crap is). Do the door frame with no cuts from door window edge till the "A" pillar door lower edge. After doing
this I had no more water leaks.

Thanks,

Patrick

[964Cab]

I too have water leaking into my 964 and the 993 Cab through the door weather striping
and getting too close for comfort to the $$$$ DME brain under the driver seat. Here is how I cured the
water leak. I bought 3/4" foam weather striping from Frost King (sold at every Home Dept, etc) It's
3/4" wide foam striping and 3/16th thick and comes in 17 feet rolls. Part number Frost King V4498. Take it and
use weather strip glue to secure all along the edge of the door frame (not the body frame where the stock factory rubber crap is). Do the door frame with no cuts from door window edge till the "A" pillar door lower edge. After doing
this I had no more water leaks.

Thanks,

Patrick

PS: Upon reading some more Porsche service tech manuals that the dealer factory techs use I found that the DME relay lives and breaths as a function of what the DME "brain" does. For example, upon starting the car the DME brain sends a ground short signal to the DME relay pin # 85. Once that DME relay pin # 85 is grounded that in effect closes another coil in the DME relay that sends a positive + voltage to pin 37 of the DME "brain" unit. That + positive voltage keeps the DME brain firing pulses to the ignitor system that feeds the coils that feeds the twin dist. Sooooo, if that DME relay is hiccuping in any way then the DME brain gets messed up and that in turn screws the engine performance up big time !

Above noted "hicupping" could be caused by any number of things such as false positive air bag signal, false positive anti theft signal, etc. In effect the DME "thinks" the car has crashed and kills the + voltage signal to the "brain" and also kills the fuel pump. Also, the DME "thinks" the car has just been stolen and disrupts that positive + voltage to pin 37 of the "brain" and shuts the fuel pump off as well as halts the firing of DME brain sequential pulses to the ingnition "ignitors" that feed the coils. So many false positive signals cause so many problems that the DME brain just halts operation.

Patrick

PS: I'm curious if an after market ignition system (MSD ?) could provide a more reliable sequental sync pulse to the ignitors so they can feed the coils and the twin dist in a fashion which could provide more enjoyment for us 964 owners (no offense Porsche).

[Jim T]

Originally posted by mwbaum
Smatta Wayne?? Doncha wanna sell more relays already??

Definately. Wayne, you should take some relays, paint them red, add $25 to the price, and sell them as "performance" relays! I GUARANTEE you that you will have nothing but satisfied, happy customers who will readily testify to smoother idle, 15-20 more hp, "freer-revving," etc. etc.

Its a no-lose situation!

[Superman]

My butt-dyno confirms an 8hp increase each time I wash the car. Runs smoother and quieter too.

[ebradway]

A yellow stripe and a 'type R' sticker on the DME really kicks...

I need to look at that board again and see where to bypass the alarm system. The PO of my car never even knew it had an alarm - and didn't have the key. So all it does is add more stuff to fail.

A failing DME can have two possible symptoms: the random dying because of circuit faults cutting power to the fuel pump and o2 sensor or mis-mixed fuel/air because the voltage sent to the 02 sensor. You keep a spare because of the first symptom, but swap with new because of the second.

I'll get back later with how to bypass the alarm.

[964Cab]

Hi Ebradway:

I too didn't know I had an alarm system until one day last week I mistakenly set the alarm off when I was figeting with the door key. I re-opened the driver door with the key and the alarm went off.

I look at the wiring diagram shows that the ignition lock unit (goes to a terminal connector on the steering column in the luagge area) has to supply + volts to the DME at DME relay pin # 86 of my 964. Now if that + volatge is intermittent due to a false positive alarm fault the DME shuts down and kills the DME brain and fuel pump as well. It "thinks" the car is being stolen or has crashed ! Also, I have an aftermarket radio and that factory radio alarm wiring is just hanging under my dash and I have no clue what to do with it (be nice).

Thanks,

Patrick

PS: Car still intermittently looses high volatge to all the plugs at once !