Never ending fuel pump drama

[Murray_Calavera]

The car hates me. I'm 100% sure it's become sentient and actually hates me. 

Cold start this morning, perfect. Cold start after finishing work, perfect. 

Didn't go straight home, parked at the shops.... go back to the car for a hot start, nothing. Pump doesn't prime. 

Currently waiting for the car to cool down, fingers crossed it starts again once it's cold again. 

[Murray_Calavera]

So I decided to return to the parking lot with tools and a fuel pump thinking what else could it be at this stage?

Umm yeah, really didn't think this was going to happen so soon (or hopefully at all).... Didn't even get 2 weeks out of the Taarks bulkhead terminals. Rated to 40 amps apparently. I'm running a 25 amp fuse in that circuit so that's interesting. 

[GTSBoy]

Yeah, see, I think I've worked out what is happening here.

You (and other people) buy these massive "40 amp rated" pass throughs, and use fat wires, then make the connection between them using a ring crimp that has a much smaller cross section in that piece in the neck between the crimp and the ring terminal than any other part of the system. And it gets hot there. And because it is very well connected to the passthrough....that gets hot too.

We probably need to be using more chunky wire terminations.

 

[Murray_Calavera]

@GTSBoy

Makes sense. 

I'm guessing we'd have to use a cable lug, its tricky finding lugs suited for wire smaller then 0 gauge, but this looks like it might work?

https://au.element14.com/panduit/lca10-10f-l/terminal-compression-lug-10-10awg/dp/2944903?CMP=KNC-GAU-GEN-SHOPPING-PLA-PMAX&mckv=_dc|pcrid||pkw||pmt||slid||product|2944903|pgrid||ptaid||&gad_source=1&gclid=CjwKCAjw4ri0BhAvEiwA8oo6F3hR5QEQH43vntEUad27nYgwWzBmgdiYNqiQLBMe3C9pkkncxgv3oRoCciEQAvD_BwE

Just need to do some surgery on the bottom of the bulkhead as even fitting the tiny ring connectors space is tight. 

But now that I'm in this situation, do I buy another stock bulkhead and give the bigger lugs a go or do I just drop $1000 on a frenchy setup? Hmmm.

[Murray_Calavera]

This might be another option too if the surgery allows for a straight lug to be used. Jaycar rates them at 30 amps - 

https://www.jaycar.com.au/6mm-non-insulated-eye-terminal-4mm2-pack-of-8/p/PT4932

[GTSBoy]

I'd try the Jaycar one first. Don't even bother with replacing the passthroughs, as they will still work provided the melting is only minor and they still seal and insulate. You can always go back and replace those later.

After you've put the new crimps on, you just have to make the pump work for a bit, then go back there and laser gun the terminals (or touch if you're brave enough) to see if they are sufficiently cool to relieve your concern. You might want to gather a data point with the old setup first, to see how hot they get now.

[Dose Pipe Sutututu]

I usually rip off the insulation off those terminals and crimp them down without it (to get a better crimp) then heat shrink over the top.

Man, just PWM your fuel pump and enjoy life lol.

[TurboTapin]

Those yellow ring terminals are usually UL rated to 48-50A, and the blue ones to 30A. 105C. I doubt that's your issue. What's oozing out of the top of that nut? Is that a nylon lock nut?

Does your bulkhead fitting have enough landing for your ring terminal to make good contact on both sides? Bad contact = Heat. Also check your crimps. Once crimped, confirm it's done properly by trying to pull the wire and ring apart. It should not budge. 

[GTSBoy]

56 minutes ago, TurboTapin said:

I doubt that's your issue.

Bah.

That bit there can't be more than 0.5mm thick, and only about 6mm wide. That's only 3mm2 XS area**, which is a lot smaller than any other part of that circuit. (Although, granted, the wire inside the tank on that blue ring crimp looks a lot smaller than the wire outside!). The equivalent neck on the blue crimp has to be even smaller.

**And yeah, I realise that 3mm2 should be able to carry ~30A. But something has to be the weak link here, and that is the narrowest part of the whole thing.

So, unless the crimps are bad and causing high voltage drop, the next most likely suspect has to be the necks. The face area of the ring onto the passthrough's landing would have to be multiples of the neck's XS area. So long as they started out tight they (the ring contact face) never should have been the source of the heat. Same with the suspicion of the possible nyloc. If it started out tight, it shouldn't have ever gotten hot enough to melt the nylon, loosen and then cause the heat that melted the nylon. The circular argument has to start elsewhere.

If it were me, I'd be thinking about some other sort of passthrough that doesn't require the nuts and ring terminals for conductivity. If there were passthroughs that had direct crimps on both sides, they'd be single use, but they would have metal to metal interfaces only at the crimps. Would be solid metal between the crimps.

Edited July 11 by GTSBoy

[TurboTapin]

37 minutes ago, GTSBoy said:

Bah.

That bit there can't be more than 0.5mm thick, and only about 6mm wide. That's only 3mm2 XS area**, which is a lot smaller than any other part of that circuit. (Although, granted, the wire inside the tank on that blue ring crimp looks a lot smaller than the wire outside!). The equivalent neck on the blue crimp has to be even smaller.

**And yeah, I realise that 3mm2 should be able to carry ~30A. But something has to be the weak link here, and that is the narrowest part of the whole thing.

So, unless the crimps are bad and causing high voltage drop, the next most likely suspect has to be the necks. The face area of the ring onto the passthrough's landing would have to be multiples of the neck's XS area. So long as they started out tight they (the ring contact face) never should have been the source of the heat. Same with the suspicion of the possible nyloc. If it started out tight, it shouldn't have ever gotten hot enough to melt the nylon, loosen and then cause the heat that melted the nylon. The circular argument has to start elsewhere.

If it were me, I'd be thinking about some other sort of passthrough that doesn't require the nuts and ring terminals for conductivity. If there were passthroughs that had direct crimps on both sides, they'd be single use, but they would have metal to metal interfaces only at the crimps. Would be solid metal between the crimps.

I've used the same style crimps on 600VAC, well above 25A. When I'm writing a bill of material for a design, I base it on what it's rated for and nothing more. If I measured every neck of every terminal I used, I'd get fired and presumably sent to an insane asylum. You're over thinking it. Now this only applies if he's using a name brand terminal and not a ching chong meow ali express special. My Radium hanger uses much smaller bulk heads and ring terminals and they're cold to the touch. 

As for my question concerning the nylon lock nut, crimps and if the ring is properly landing on the bulkhead, what I was getting at is if the bulkhead landing area is small and the ring terminal is oversized for the stud and not making a good contact, it could cause the heat with a nylock. As for the heat, if that's in fact a nylon lock nut and the nylon liquified which is what it looks like in that picture, it took some serious heat! Something was poorly done here, perhaps loose nuts or the crimps themselves. Looking at those crimps, it looks like he didn't use "Insulated terminal" crimpers as there's no indent anywhere to be seen. 

Edited July 11 by TurboTapin

[Murray_Calavera]

3 hours ago, GTSBoy said:

see if they are sufficiently cool

Sounds good, what temperature range would you expect if everything is healthy?

1 hour ago, Dose Pipe Sutututu said:

just PWM your fuel pump and enjoy life

It's coming I promise, it just has to wait a little bit first. I don't really have the time to set it up at the moment. 

1 hour ago, TurboTapin said:

Those yellow ring terminals are usually UL rated to 48-50A, and the blue ones to 30A. 105C. I doubt that's your issue. What's oozing out of the top of that nut? Is that a nylon lock nut?

Does your bulkhead fitting have enough landing for your ring terminal to make good contact on both sides? Bad contact = Heat. Also check your crimps. Once crimped, confirm it's done properly by trying to pull the wire and ring apart. It should not budge. 

Yeah, surprised me too. Everything I read rated those ring terminals quite high. Perhaps the issue was with me bending the ring terminal about 90 degrees? 

Yep, the Taarks kit comes with nyloc nuts. This photo of the kit shows it pretty well. 

I'm not quite sure what you mean by having enough landing for the ring terminal though? 

As for the crimps, they are solid. I've crimped a million connectors in my car before, all the sensors, my dash, the quick connect terminals on my battery etc. 

40 minutes ago, GTSBoy said:

Although, granted, the wire inside the tank on that blue ring crimp looks a lot smaller than the wire outside!

Correct, but both are rated for the use. The wire inside the tank is 12 gauge Tefzel wire, it's rated for 25 amps continuous. The wire on the outside is 25 amp Jaycar wire. I only use Tefzel when I don't expect the change the wiring any time soon. As I expect to change the wiring outside the tank in the near future to setup PWM, I don't want to waste Tefzel wire. 

10 minutes ago, TurboTapin said:

using a name brand terminal

The ring connectors would be from either Supercheap or Jaycar.

13 minutes ago, TurboTapin said:

perhaps loose nuts

I triple checked it, the nuts were tight on both side of the bulkhead. 

[GTSBoy]

22 minutes ago, Murray_Calavera said:

Sounds good, what temperature range would you expect if everything is healthy?

Not more than 50°C. Anything you can't touch is putting up more of a fight than it should be.

[Murray_Calavera]

Just stumbled on this, sounds interesting - 

https://www.efihardware.com/products/3194/submersible-bulkhead-terminals-hex-head

"Hex cup versions enable Insulating sealing cups to be torqued up against each other ensuring a tight seal eliminating bad resistance resulting in excessive heat from high amperage."

When I installed the Taarks kit, I tried my best to tighten down the white plastic insulating , but they are round and threaded. So unfortunately the nuts don't cause them to clamp down due to the thread and I couldn't grip them very well to torque them down. I'd say they are only finger tight each side. The nuts however were torqued down well on each side. 

Just reading the Efi hardware site further, it just gets worse. So essentially they are saying the Taarks kit is only rated to 10 amps and should not be used on a plastic bulkhead. 

They recommend this connector that has the added flange on it for a plastic bulkhead - 

 

 

[Murray_Calavera]

44 minutes ago, TurboTapin said:

if the ring is properly landing on the bulkhead

Any chance you are talking about the flanged area in the picture above as the 'landing on the bulkhead'? If so the Taarks kit doesn't have that.

 

46 minutes ago, TurboTapin said:

if the bulkhead landing area is small and the ring terminal is oversized for the stud and not making a good contact, it could cause the heat with a nylock.

The ring terminal is slightly larger then the stud, I would not describe those two as making good contact with each other. My assumption was, the good contact comes from the 2 washers on either side of the ring terminal making a good connection, and the washers make a good connection with the nut, which makes a good connection with the stud. I'm guessing this was an incorrect assumption? 

[Kinkstaah]

20 minutes ago, Murray_Calavera said:

Just stumbled on this, sounds interesting - 

https://www.efihardware.com/products/3194/submersible-bulkhead-terminals-hex-head

"Hex cup versions enable Insulating sealing cups to be torqued up against each other ensuring a tight seal eliminating bad resistance resulting in excessive heat from high amperage."

When I installed the Taarks kit, I tried my best to tighten down the white plastic insulating , but they are round and threaded. So unfortunately the nuts don't cause them to clamp down due to the thread and I couldn't grip them very well to torque them down. I'd say they are only finger tight each side. The nuts however were torqued down well on each side. 

Just reading the Efi hardware site further, it just gets worse. So essentially they are saying the Taarks kit is only rated to 10 amps and should not be used on a plastic bulkhead. 

They recommend this connector that has the added flange on it for a plastic bulkhead - 

 

 

These are the ones that I melted.

You need the FPR kit, nothing else will work. :p

[Dose Pipe Sutututu]

1 hour ago, GTSBoy said:

unless the crimps are bad

I nearly feel this is the issue, those insulated terminals are aids to crimp correctly.

[Murray_Calavera]

3 minutes ago, Kinkstaah said:

These are the ones that I melted.

Those exact ones, with the hex head on the plastic insulators and the metal flange? 

[Kinkstaah]

Okay it seems there's a slight design difference there with the hex which will do nothing at all. The ones I had were the EFI Solutions round ones. I was able ratchet the nuts with huge torque and they were flat and FIRMLY into it.

You could say they became part of the lid, given they physically melted into it. It doesn't get any more firmly attached than that!

The heat generated is enough to melt the plastic insulators. I agree, you would think Walbro would use wiring in their own pump with enough gauge to handle the heat their own product creates, but it just doesn't seem to be the case long term. I suppose it's possible that mine only shifted towards the end of their 10 year life before melting through, but that said... it can shift again, a nut can come loose again.

The solution is a different type of plug, that is held in mechanically and can't shake loose and self-create a bad connection.

[Murray_Calavera]

Just looking at the instructions for these Efi hardware terminals, is anyone able to explain to me why the diameter of the hole must be specifically between 8.1 and 8.2 mm? If the post is 6mm, and you drill say a 7mm hole, how could this possibly cause an issue? 

In any case, the more I read these instructions, the more I think I want to give them a go. 

https://www.efihardware.com/download/237/866-033_M6_Bulkhead_Terminal_High_Amp

[GTSBoy]

1 hour ago, Murray_Calavera said:

is anyone able to explain to me why the diameter of the hole must be specifically between 8.1 and 8.2 mm? If the post is 6mm, and you drill say a 7mm hole, how could this possibly cause an issue? 

Hypothetical hat on.... I think they want to make sure that it is no more than 8.2mm to ensure that the 14mm hex electrical isolator washer thing stays outboard of the hole, so the o-rig on it seals properly, and they want as much room as possible between the passthrough bolt and the plastic lid, so that it can't touch the plastic and heat it. Even assuming that the steel should run cold enough to not be an issue, they are probably applying additional paranoia in case of the sorts of crimp/terminal problems we're thinking might be at play in your situation.