Media | Articles
Power Up: Two Valuable Relay Troubleshooting Tricks
Not long after I began contributing to Hagerty eight years ago, I wrote a detailed four-part piece on electrical relays and how they work in cars. You can still find the articles here (Part I: Why We Need Them; Part II: DIN Numbers and Different Types of Relays; Part II: Troubleshooting, and Part IV: The Solenoid).
The main thing you need to know is that a relay is a remote-controlled switch that uses a small amount of electrical current to energize an electromagnet that pulls two electrical contacts together, allowing a much larger amount of current to flow, and turn on a power-hungry device like an electric fan. Most relays are of the four-prong, single-pole-single-throw (SPST) variety, and have the following four DIN-standard terminal numbers:
- 30 is the high-current 12-volt supply from the battery.
- 87 is the high-current output of the relay to whatever you want to turn on (e.g., an electric fan).
- 86 is the low-current voltage from a switch that turns on the device.
- 85 is the low-current ground.
This is depicted in the professionally-rendered diagram below. From here you can see that, to use a relay to turn on the electric motor labeled “M,” you ground the negative wire of the motor, connect the positive wire to the relay’s “87” terminal, and connect the relay’s “30” terminal to battery ground with a fuse. That completes the high-current side of the wiring. You then wire the relay’s low-current side by connecting 86 to a source of voltage with a fuse and a switch, and ground 85. When you flip the switch, the current flowing through the electromagnet energizes it, pulling the contact from 30 to 87 closed, and causing current to flow through the fan and to ground. Pretty cool, eh?

From here, you can see that, when people talk about “jumpering” a relay, what they mean is pulling the relay out of its socket, identifying the terminals in the socket for 30 and 87, and using a jumper wire (ideally one with male connectors at both ends) to connect them, thus bypassing the relay entirely and causing current to flow to the thing the relay is supposed to turn on. Note that the terminal numbering is usually embossed on the bottom of the relay, so you need to write that down on a small slip of paper and put it next to the relay socket to get the numbers of the terminals to jumper.

There are sometimes small complicating factors. One is that on simple 4-pin SPST relays without internal diodes, current can flow through them in either direction, so you may find that they’re wired on the high-current side with 87 and 30 switched, or on the low-current side with 86 fed 12 volts, and 87 grounded. Another is that in some applications, like horns, the switch is on the ground side of the relay, not the 12V site.
With all that context, I can tell you the following two tricks I recently employed in troubleshooting a relay circuit.
I was working on the FrankenThirty, the salvage-titled 1988 BMW E30 325is that I’m still sorting out. The car has a two-speed electric fan that functions as both a condenser fan for the air conditioning and an auxiliary cooling fan for the radiator. In the first situation, the low-speed fan setting is triggered by the a/c “on” switch; in the second, both low and high are toggled on by a temperature sensor screwed into the radiator. The high speed worked, but the low speed did not. Each speed is controlled by its own relay. The online diagnostics in E30 user forums are excellent, and implicated the low-speed relay labeled K1. I verified that there was voltage at terminal 30, jumpered it to 87, and the fan whirred to life.
So the problem had to be on the relay’s low-current side. I verified that when I hit the a/c switch and found that terminal 85 had current (remember what I said about 85 and 86 sometimes being swapped). So either the relay itself was bad, or there wasn’t a ground connection on 86. I swapped the relay with the working one for the fan’s high-speed setting, and it made no difference. I then set the multimeter to measure resistance, connected it between terminal 86 and ground, and did not measure continuity. So terminal 86 wasn’t grounded.
Gotcha.
Because relays are almost always plugged into sockets, and because these sockets are often recessed inside a fuse box, there’s no obvious way to test the terminals with a multimeter while the relay is plugged in.
Unless, of course, you use the following trick.
First, a bit of terminology. The flat terminals on the bottom of the relay are often referred to as “male spade connectors,” but a spade connector is really a two-prong-style connector that goes around a stud. The flat male connectors (and the female ones that receive them) are technically called “quick disconnect connectors,” which is an awkward term. You’ll see them listed on Amazon.com with both names.
Okay, on to the trick. What you’re going to do is make a set of extension wires. Take four lengths of wire and cut them to the same length, about a foot. Two of the wires should be of a decent thickness, like 16-gauge, for the high-current side of the relay. The other two for the low-current side can be thin-gauge.
On one end of each wire, crimp a female quick-connect terminal. Crimp a male terminal on the other end. Slide the female ends onto the terminals on the underside of the relay. Using that piece of paper you wrote up as a guide (shown above in the “jumpering” photo), carefully plug the male terminals into their corresponding locations in the socket. Be absolutely certain you get this correct. Due to the orientation of its pins, a relay can only fit one way into the socket. But there is one way to plug in the extension wires correctly, and eight ways to do it wrong.
Now, lay the relay somewhere that you’re certain the live voltage connections aren’t going to accidentally ground out. Sliding a piece of cardboard under it is a good idea. I didn’t. Do what I say, not what I did.

Congrats. You now have a test bed where you can easily probe the relay terminals, verify their voltage and ground status, and if necessary, introduce voltage or ground. In my case, this enabled me to ground relay terminal 86. The fan then roared to life on the low-speed setting.

Obviously I couldn’t drive the car this way. This was strictly a temporary troubleshooting measure enabling to me to figure out the source of the problem. I searched online and learned that the low-speed fan is grounded on a stud located way up under the dashboard. I found it and examined it. And it looked fine, not the corroded mess I was expecting. Nonetheless, I pulled it apart, cleaned it, and reassembled it. Unfortunately that made no difference.
So, how do you ground the leg of a relay that you can’t access when it’s plugged in? I was faced with having to pull up the fuse box and access the ground wire underneath and tap into it, but I hated to use such a big hammer to solve the problem when I still had so much other work to do on the car.
And this leads me to tip number 2. Okay, you can call this one a kluge rather than a tip, but it worked. There exists what are sometimes called “piggyback quick connect adapters” (or, incorrectly, spade adapters) that are like a little two-on-one, quick-disconnect connector power strip. They have a female connector that slides over a male connector and presents two males that you can plug things onto.

So the trick here is to take four of these, straighten them out, and stick them on the underside of a relay. Due to space, you probably can’t connect extension wires to all of them, but you can orient the adapter so you can connect an extension wire to one of them.

And that’s what I did. But first I verified that the fuse box’s weather cover would still snap shut with the relay sitting a bit higher, and it did.

Then, by carefully threading the ground wire between the cover and the side of the fuse box, I was able to snake it out and attach it to a grounded lug in the engine compartment. Yes, I do have some concerns about the weather sealing of the fuse box with the wire in place, but for now it will allow me to continue to drive the car as I sort it out.

So there you go. Not one but two relay-circuit troubleshooting tricks. Knowing how useful they are, the most amazing thing is how I’ve gotten through nearly 50 years of wrenching without having them in the mental toolbox until this week.
***
Rob’s latest book, The Best Of The Hack Mechanic™: 35 years of hacks, kluges, and assorted automotive mayhem, is available on Amazon here. His other seven books are available here on Amazon, or you can order personally inscribed copies from Rob’s website, www.robsiegel.com.
Excellent solution. It really embodies the “don’t get it right, just get it running” mantra.
No disrespect Colton, but you must not know many amateurs. I know guys who would have twisted some 20 gauge or speaker wire right into that relay socket, twisted the other end onto any nearby body bolt and called it good. These are fellas that think electrical tape is “fancy” and best used as first aid equipment. I don’t claim to be one of them, but I may be related.
I know a lot of em. I’ve fixed their handiwork.
But Rob does my approved method, which is doing the wrong thing the right way. The other guys are just creating fire hazards.
I think you meant to say: “You then wire the relay’s low-current side by connecting 86 to a source of voltage with a fuse and a switch, and ground 85.” At least, that’s what your professionally rendered wringing diagram is depicting.
Whoops! Thanks for catching that. I’ll get it changed.
I had a “similar” thing happen. I had a circuit that was intermittently failing. I was probing around the relay socket for that circuit and I noticed (probably more like stumbled upon) that one of the female connectors was moving down into the slot. The latching tab on the connecter had pushed back into the connector thereby not securing the connector into the “slot”. So, the male connector was not getting full insertion. Yeah, I said that… Anyhoo, I pulled the relay box up and luckily there was enough slack to flip it over on its back. Or is that on its face… As I was already that deep into the fix, I decided to tug on each wire and found a couple of more loose connectors. Pulled them out, pulled the latching tabs out on the connectors, pushed them back into their slot. Nice and tight again!!!
Reading your article, which I enjoyed (thanx), gives more proof to how much I like working on wiring. Although I was raised in a family that primarily owned VWs, I started wrenching on Japanese beer cans and now work on probably too much. But DIN is DIN regardless what brand you’re working on. So, some of your articles are, for me, a good refresher course mixed in with a good story. And the end result? In your case, in the least you’ll have your FrankenThirty in decent daily-driver shape. Win-win!
I like the quick connect solution, I’ll try that one. The render was very well done too!
Rob, I think you also meant to say connect the 30 terminal to a fused power or positive battery source.
That first jumper wire reminds me of my water cooled VW days when I kept an 8 amp inline fuse holder with a male quick disconnect on each end to jumper the fuel pump relay.
There was a factory tool that was a relay housing with a switch on top but the DIY was cheaper
It’s easy to diagnose these relays, but much harder if you have a Ford with the Integrated Relay Control Module. You can always use the voltmeter to directly measure power to the relay terminals. Is terminal 86 supposed to be hot at all times under all operating conditions? I’m guessing the answer is yes, which means terminal 85 is not connected directly to ground. I’m not familiar with foreign cars, but given that it’s a 1988 it probably has a computer that controls the powertrain and it works not by supplying power, but by grounding out circuits. If you dig deeper into this rabbit hole you’re going to find that the relay is triggered by a transistor, or maybe three if you’re dealing with the AC relay in a Ford, Mercury, or Lincoln. It might also be possible to diagnose the problem using on-board diagnostics, and I would try that to see what kind of result happens. I have found trouble codes even though the check engine light wasn’t on. The fan might turn on with the AC, or it might not, if the computer thinks the engine is not at operating temperature – I don’t know enough about that, but before I do anything else I would diagnose the engine coolant temperature sensor.
I found a set of “breakout leads” online that have the male/female extensions like your first tip, but also with a multimeter test lead plug in the middle. They work great for troubleshooting relays.
Deja Vu all over again… Rob, you brought a smile to my brain as I remembered my beloved/behated ‘79 VW Rabbit, AKA “The Pride Of Westmoreland”. I got quite adept at jumpering the connections on the self-destructing fuse panel. No lie, I kept a roll of wire and a copy of the electrical schematic in the glove box for emergency roadside repairs. Legend was that rust-clogged fuel lines overloaded the fuel pump, melting the circuit board traces. But those were just the ones that were left after rainwater got in and corroded the others. A junkyard donor fuse panel mostly fixed the problem, and years later I handed the new owner two extra “just in case” fuse boxes to go with his purchase.