Main Relay failure
 

Larry Koh created this post in the Maxxis Cup Supermiata facebook group about his son Kyle's Maxxis Cup car. I thought it would be good info to share.
Keith Tanner chimed in that he virtually lost the Targa because of this same failure.
-Emilio

Might want to inspect your main relay.

As some of you know, Kyle Koh broke twice at Big Willow last Saturday with electrical failure. First of all, a big thank you to William Chen, Jerry Awesom, Alex Renderos, Sonny Watanasirisuk, and everyone else that helped us.

It was as if the ignition switch was just turned off while driving. Crank, no-start. Ultimately what we found was that our main relay in the engine power distribution box had failed. My first reaction was that this was a 25 year old relay that had been in use for presumably ~125k miles, so this may be the reason for failure. I called every auto parts store from Palmdale to Mojave with no luck. William came to the rescue and borrowed Jerry’s main relay, as he was done on track for the remainder of the day (another story). Kyle used the relay for qualifying (P5), but at the end of the session the same failure happened. Turns out we killed Jerry’s main relay as well.

William to the rescue again, giving his own relay after the race to Jerry so that he could drive home. Thank you again, William. New relays for both of you are on their way from Mazda. Message me your addresses, and I’ll send them later this week.

Now, what happened? I’m waiting for the relays to arrive for further testing, but after studying the failed relays and the wiring diagrams that I could get my hands on, I’ll share what I know now, so that others don’t suffer the same fate, although knowing what I know now, I could have made a modification, and Kyle could have finished the day (shhhh, don’t tell him that).

The bottom line is that continuously high current flowing through the relay caused the load side contact to heat up and melt the plastic which allowed the contact to separate enough to cease current flow, which is akin to shutting off the ignition switch. What caused this? Well, my suspicion, which I’ll verify once the new relays arrive, is the recently installed AEM 50-1200 E85 fuel pump, as that’s the only high current consumer that I can find on the wiring diagram (current actually goes from the main relay to the “circuit opening relay” to the fuel pump). Most of the direct outputs from the ECU are low current (as you might expect from a solid state component). I’ll check the actual current consumption as soon as we’re up and running again. Might need to relay control the fuel pump a different way, or look for a different fuel pump altogether.

When we couldn’t get William’s relay out of his box, we removed Alan Tseng’s. I found it was quite hot when I removed it. We ended up getting William’s out, so ultimately used his for Jerry’s car, and I didn’t find his as hot, so maybe this is an NA thing? Nevertheless, if you use a fuel pump other than a bone stock one, or your relay is original, you may be setting yourself up for the same fate that we experienced. Also, it’s likely to happen over a period of time, and we just finished off Jerry’s relay which was older than ours by ~75k miles. Look at the photos, and you’ll see what I mean. Our relay wasn’t as bad as Jerry’s, but it still didn’t allow current flow through the load side.

The inspection is simple. Remove and inspect the load side pin on the bottom of the relay. If there is any sign of melted plastic, or the pin isn’t perfectly parallel to the adjacent pin, you should replace it. I would. As far as I can tell, the Denso B6S8, and the Imasen B5B4 relays are the same, although I will dissect the B5B4 when they arrive later this week and report that as well as the actual current flow of the fuel pump.

Also, if anyone has any other wiring diagrams or experience with main relay failure, or any other common failures, please comment. I’m constantly learning about our car, and knowledge is power. Just thought this information might be useful to others.

https://cimg6.ibsrv.net/gimg/www.mia...8982b68636.jpg
https://cimg8.ibsrv.net/gimg/www.mia...e912b8410d.jpg
https://cimg0.ibsrv.net/gimg/www.mia...5d82372332.jpg
https://cimg1.ibsrv.net/gimg/www.mia...de8c243395.jpg

I had the main relay die in one of my 94's right before MaTG this year. Happened right as I pulled into my driveway. I might try to find it tonight if it did not get thrown out to take a look and compare it to your pictures. The 94 has the standard VVT/Flattop/MS3 swap and nothing really else. Still on OEM fuel pump as far as I know. The car has approximately 186k.

If the circuit open relay was used simply to engage a tertiary relay for the pump and wired directly to the battery, would this reduce the current load on the main?

I went through at least 2 (maybe 3) relays last year. I now keep a spare in my car just in case. All of them showed the gap in your pictures. Since then I have changed the wiring to my COPs and made sure that all my grounds are touching clean bare metal with a little bit of copper "gasket" (not sure what its intended purpose is) in the bolt holes. I haven't had a failure yet this year, so hopefully that helped solve the problem.

I always keep a spare. And all of the NA relays I have pulled out are brown and kind of melty by the power pin.

Its on my list of to-dos to get some sort of solid state relay there. Big FET or something on a heat sink.

If someone made a plug and play version of this they would be awesome.

I replaced the main relay on my 2002 about 5 years ago with a generic 40A automotive relay, after I went through several stock relays since 2006 (one every 2 years I think). I haven't had a problem since. On the plus side, the generic ones are much cheaper than the Mazda relays and can be bought literally anywhere.

Main relay issues are well known in the Miata community, especially on NB.

My solution was to install a generic one just like Reverant. I used a gutted main relay, attached wires to the four terminals which extended outside the fuse box to a generic relay base.

YES!

I wrote about this many years ago. The first mod I will ever do to a car is to rewire the fuel pump. If you meassure voltage at the battery and then you trace the long thin 16 gauge wire to the fuse box, back to the ECU and ignition components back to fuel pump there will always a voltage drop. The solution is to use the same 12V wire that powers the fuel pump to activate a relay with a thick 10gauge wire straight from the battery to power the fuel pump.

Did some more poking around. Ordered new NA and NB relays from Mazda. Both were Denso coil type, like the one on the left.
Also ordered Standard Motor Products relays for NA and NB. The NA variant ( RY-288 ) was this solid state unit with no external markings. Anybody got any insight as to whether this is merely cheaper to make and not as good as the coil type or if it might actually work more bester?

Few other tidbits:

OEM NA relay is Denso # 056700-9000
Mazda OEM replacement relay is Denso 056700-8780 and marked "80a main relay" on the packaging

NB relay from Mazda and Standard Motor RY-667 products are the identical Denso B5B4 relay



https://cimg2.ibsrv.net/gimg/www.mia...0be0eeb1c7.jpg

The relay on the right isn't "solid state"
It's merely a much more compact off-the-shelf relay in a black plastic case, which they then made fit, with a PCB, into the miata relay box.

This is probably why my old MSM would randomly cut-out while driving...

Thanks for the info!

I changed mine around 130k miles on the chassis, not long after upgrading to a bigger fuel pump. I think the factory design is just too small of a relay and we're asking too much of it with additional load for more electronics.

I now have separate relays for the fuel pumps that get there power from the battery, and the coils/injectors/fans pull right from the alternator (with there own relays/fuses of course). I dunno how much current goes through the main relay now, but it's less than factory and haven't had a main relay fail since doing this in ~2009. The voltage drop through the factory relay was a lot with high loading.

Which is likely why the NB alternator control in the ECU outputs such a high voltage - it is trying to compensate for what it sees as a low battery voltage.

I have had this happen three times and leave me stranded. I have a DW300 pump at 60psi base so the draw is real. Factory unit lasted 5 years for me, replaced it with a unit from orielly and it died again 4 months later, and again a few months after that, and again last fall. I'm going to do exactly this though, I picked up the DW kit which appears to have everything needed unless you want to run bigger wires through the bulkhead into the tank, Savington and Aidan have been looking into this IIRC in your threads. I think and hope this fixes things.

We have not had any failures in our cars but are seeking to prevent them for the enduros we will be doing this year. One side is determining if certain replacement relays might be better than others. The other side is, as some of you have done, rewiring for separate universal relays.
Vegas is getting a DW300 and will run at least one 24hr event this year, thus the homework.

FWIW I have pulled like 3 or 4 main relays out of completely stock cars and they all have the toasted looking plastic on one pin. It probably is just an overworked relay, or a poor original design. If I were going for total reliability it would get replaced with either a generic bosch relay, or some sort of solid state thingy (if baller)

Fans, fuel pump, headlights are the biggest draws.
Fusing and relay'ing them separately would go a long way to off-loading the main relay.

(Do headlights draw through the main relay?)

Tells us about solid state relays.

A 40A generic relay works, and you can have one spare with you. I was changing OEM Mazda ones every two years, now I haven't changed one in 5 years and I have a Walbro 255 in the tank.There's no need for an SSR.

More expensive and tend to be larger because they need heatsinking.
But longer life and can do PWM'ing.