How warm should a relay get?

Question

I'm chasing a number of DTCs on our 2002 Jetta TDI (ALH engine, 3rd generation, late production, controls with 4-wire glow plugs). Most of the codes are from components that share a common power source the ECM Power Supply Relay (marked 109, called J317 if VW schematics) which is switched by the ECM when the key is turned on. We've also had leaks from the sunroof drain in the left (driver's) side A-pillar which means water gets close to the relay panel where the 109 relay is mounted.

So, I've been investigating that area and when I pulled the relay it was warm. It looked nice with no sign of water damage, but it was warm.

In the interest of actually understanding the problem I've been "playing" with the relay. Here's what I've found so far:

• The coil resistance is about 145Ω when the relay is cold and about 165Ω after the coil has been energized for 30 minutes.

• The coils pulls 62.1 mA.

• After about 30 minutes with the coil energized, but no current on the switched contacts, with an ambient temperature of 18° C (measure with a K-type thermocouple on a DMM), I read a temperature of 25° C when the relay is sitting on the thermocouple which is sitting on my desk. By feel I might be able to tell that the case is warmer than ambient, but it is definitely cooler than it was when I pulled it from the car.

• The resistance of the switched contacts is essentially unreadable - my DMM (Fluke model 73) shows 0.0Ω with an occasional reading of 0.1Ω. But, the loads on the relay are fused at 10 A, 30 A, and 10 A (ECM and injection pump, PCV heat, and engine control solenoids and the MAF sensor). At 50 A with a contact resistance of even 0.01Ω the power dissipation would be 25 W. That seems like enough heat to be a problem and enough to make the relay feel quite hot.

So, the questions:

• How warm should a "normal" relay be, assuming a 100% duty cycle and a reasonably high current? I assume some heating from the coil could be normal, but corroded contacts would also produce heating and one explanation for the codes could be the relay dropping out.

• Should I replace it, just because? They are cheap.

Codes for anybody who's interested:

• 17946 (P1538) - Fuel Shutoff Solenoid (N109) - 35-10 - Open or Short to Ground - Intermittent

• 17849 (P1441) - EGR Valve (N18) - 35-10 - Open Circuit or Short to Ground - Intermittent

• 19561 (P3105) - Valve for Intake Manifold Flap (N239) - 35-10 - Open or Short to Ground - Intermittent

• 16629 (P0245) - Solenoid Valve  for Boost Pressure Control (N75) - 35-10 - Short to Ground - Intermittent

• 16600 (P0216) - Commencement of Injection Valve (N108) - 35-10 - Circuit Malfunction - Intermittent

Answer 1

I agree with everything you have to say here. The relay is going to create some amount of heat. How much heat is really unknown. I believe relays will cut out if they get too warm, but I'm unsure of how much heat buildup would cause it to do such.

Considering the price of a relay, I'd highly suggest just changing it out. This is especially true if you are seeing any kind of corrosion on the terminals of the relay or in the socket. If you see it in the socket, you'll want to do something with it as well. If you don't you'll find you haven't solved your issue or it will cause problems later down the line (problems of the same sort). How to clean the socket is a whole different issue, one with which I've not dealt with before. I'm sure you can figure something out, though. If you change out the relay and it doesn't fix your issue, it just means you have a good spare, which is never a bad thing. Relays are fairly common between brands of vehicles, so it could probably be used in more than just the one you're replacing it in.

Answer 2

Regarding the heat up of your relay:

If the resistance of 145 or 165 Ohm is correct - and this is a reasonable value - the coil itself will have a power loss of around 1 Watt (See details below).

That power loss translates into heat dissipation. While 1 Watt of heating power may seem little, it is enough to noticeably increase the relays temperature.

A sample calculation I did showed that a 1 W heat source connected to a tiny heat sink will heat the sink to 33°C in an environment with 18°C air temperature.

So even on a correctly working relay, without any heat generated from current flowing through the load side of the relay, there is some heating going on.

Details:

• ~ 0,87 W at 12V (165 Ohm)
• ~ 1 W at 12V (145 Ohm)
• ~ 1,2 W at 14V (165 Ohm)
• ~ 1,35 W at 14V (145 Ohm)

Answer 3

The 109 relay has been the cause of stranded VWs. I have had the original ones open up, and not provide power leaving me with a stranded car. This was early in the life of the car. I think the last batch I got were about $6 so I would replace it.

The failure modality in all the failed ones I had were cold solder joints for the contacts. There is moderately high current through this relay, and the solder tended to crystallize making a cold solder joint. I have reflowed that solder joint and restored operation of the relay, but seemingly the newer ones tend to fail less.

At the very least, put a new relay in, and disassemble the old one, and reflow the solder around the contacts. Then keep the refurbished relay in the glovebox in case you need it someday. Mysteriously, keeping a backup relay reduces the odds of having the installed one fail.