Jfrahm

I just had a no-start in my 1987 928S4 which turned out to be the typical bad fuel pump relay. I didn't have a handy spare that matched exactly so I pulled the cover off to see what was what.
Sadly I saw that the contacts were not squared up, looked like a manufacturing defect, and the loss of contact area appeared to have caused the contacts to burn. Some emery cloth and a little bending sorted it out and it's working. I recall I had the same situation with an AC relay in my 968.
It makes me wonder if there is a better relay made by a robot or something that'd be a better replacement. I could also pull the horn relay and see if those contacts could be improved and swap them around, this is probably reasonable as the OEM one did make it almost 40 years.

Alternatively, maybe a Panasonic CB1A-M-12V or ‎HELLA 933332151 would be a better choice, and I also see some similar relays with a telltale LED. Maybe not a bad upgrade if the parts are of reasonable quality.
Thoughts?

Cheers, -Joel.

The Forgotten On

I used relays made by Gebe. They're made in Germany and fairly inexpensive. You just need to figure out which type you need. With a resistor or with a diode. They make both and ones without either as well.

https://industrialtecsupply.com/collections/gebe/style_4-terminal+volts_12v

Mrmerlin

As a side note most relay damage occurs when they are subjected to under voltage operation
With a low battery the relay doesn’t get enough voltage to close thus the contacts arc and create a poor contact surface
Over time the condition gets worse till the relay won’t pass current

Bottom line keep the battery charged use a charger to charge the battery every few months not a tender

disconnect the ground strap if the car will sit longer than a week

DHS928

I am trying the relays with LED’s will see how the work.
I do not how long they will last but it seems like a good idea.

Jfrahm

Thanks, good advice re: the battery. Undervoltage kills a lot of stuff.

Good catch on the resistor or diode, these relays appear to have a resistor aboard as per the diagram. I didn't notice one at first when I was repairing mine but did see it later. The relay is 77 ohms across the coil pins. The LED equipped relays look like a good troubleshooting aid, interesting. In the near term I decided to clean and square up the terminals in the horn relay and swap them.

While I was at it I swapped the ABS relay with a spare and that resolved an ABS fault nag. Pulling the bad relay's cover off, I see that it too is poorly made and the issue is probably cold solder joints with voids in them. This module is actually a board with a small relay, a resistor and a big diode.

FredR

The logical upgrade would be to install relays using solid state technology rather than antiquated electro mechanical devices just a matter of whether one would consider whether the upgrade is worth it. Electromechanicl relays are very cheap and reliable and not an upgrade I would consider but whatever floats one's boat.

hans14914

Fred, It is more difficult than you would think to find ISO relays with both polarities in relation to trigger and flow for the output.

The 928 uses triggers in both direction as well as with the switched load.

I have been using a good quality relay that I currently can only source in smaller quantities from Fueltech. It has a 40a armature and has a fuse socket. I use them in the LH and EZK positions on my aftermarket ECU installations to provided fused power. Oddly enough, there is no circuit protection for the LH or EZK in the stock configuration.

Jfrahm

For the fuel pump, I think a solid state "relay" is a somewhat questionable idea. In an accident or fire, it's important that this relay open up and stay open. Even a short engagement can move a LOT of fuel. A circuit that replaces the relay might be more susceptible to false activation due to electrical noise. Better the devil we know, IMO. If someone does build one I suggest having it be smart enough to look for a strong, consistent signal and for and for maybe 250ms so it might even be less likely to false due to some torn wiring arcing or the like.

In the past I suggested an LH relay with some built in power conditioning. I still like that idea but it might be redundant if the LH has been rebuilt. The goal would be to protect the LH from a loose battery cable, bad voltage regulator, mean battery booster, etc.

UKenGB

Why do you say that when a battery tender is specifically designed to keep the battery in optimum condition for however long it is connected?

In my experience that is exactly what they do and are better for a battery than allowing it to get low then fully charging it. Lead acid batteries like to be kept charged. Little and often. Why do you think a tender is bad for a battery?

FredR

Hans,

Always good to hear from you on this list. As an FYI I classify my knowledge of electronics as "dangerous"!

I have no clue whatsoever about what "is or is not" available in terms of the various combinations of relay sub options. I consider what Porsche provided as being more than adequate but in some respects somewhat puzzling. When I was trouble shooting an engine management problem a few years ago I developed a SLD for the system as my mind goes blank jumping around from page to page of wiring diagrams - once I have a system SLD in front of me I find trouble shooting so much easier. Indeed I also noted what appeared to be a lack of a fuse in the LH/EZK sub systems but figured I must be missing something so thanks for that snippet of info!

I was puzzled as to why our 928's seemingly did not have a "DME" relay like other Porsche's of that era and for whatever reason it turned out it was a "Motronic versus Jetronic" thing. Our 928's having the Jetronic system were able to switch the fuel pump on via the secondary board in the LH unit that invariably goes "**** up" as has happened to me twice on two separate LH units over the last 24 years of 928 ownership. The second failure caused me to look into whether the secondary board failure could be ignored. Of course the beauty of our system is that if the engine stops as in the huge accident I had back in 2005 in my late 90S4, the fuel pump immediately stops and thus the risk of a fire is reduced considerably. After the impact I was in another world for a short period of time and when I came round everything was "quiet and serene". Getting out of the car with a fractured L1 vertebra and a whiplash injury in my neck was "challenging" to say the least. That I survived was in no short measure due to the Porsche design. Despite hitting a parapet nearly head on at around 70 mph due to some **** on a mobile phone is what it is and why I am reluctant to modify such systems or recommend such to anyone.

That being said I know there are plug and play relays and as I understand the critical relays [type 053] are simple SPST items and should be readily available- quite happy to take your guidance on such. Our friend who raised this thread asked a simple question as to whether there were better options and there are but what he did not ask was whether it is meaningful to do so. When I started our on my career back in the mid 70's every process plant in he oil patch relied on relay logic for process safeguarding and within a reltivel few years every new build and most revamps deployed SSL. I suppose we could convert to a DME type system and thus ignore the LH common failure mode but I suspect the time and effort to do so is just not worth it. I have three LH units two of which are upgraded/repaired by JDS Porsche and I also carry a set of 053 type relays and two relay jumper cables in the glove box.

The only other relay I would consider worth upgrading is that headlamp monstrosity- that might be a worthwhile project for a bright and talented chap like yourself . When my headlamp relay crapped out I simply got a spare from my 90S4 bag of relays. If that one takes a dump I will be screwed until such time as I stump up US$200 to brother Roger.

hans14914

Standard electromechanical relays are essentially bi-directional. They dont really care which direction the current flows through the coil. Just one side has to be positive and the other negative. Similarly, when this coil is energized, it just flips a switch inside bridging the pole to the normally open contact.

On a solid state relay, polarity is critical for both the trigger and the load. They can only supply +12v on the output, or allow connection to ground based on the internal design and if its a PNP or NPN style transistor doing the physical switching. Similary, the trigger input is determined by construction to be either trigger on a positive or negative input.

There are several ISO format relays out there, but the more common ones, say the Hella H41773001 only accepts a negative trigger input, and will only supply +12v out to the downstream load.



Examination of the totality of the relay system is required to see if it will function in that manner.

MRS offers several ISO footprint relays, and have offer both negative and positive triggers. However, they use the supplementary 87a terminal for current feedback and fault detection, so you have another live pin in the CE panel going into an unintended orifice. I doubt anything would short out or make contact, but insulating that terminal (or removing it) would probably be advised. They are also a pain to get, even though they now have a sales office in Ohio.

I will need to dig into the schematics of the airbag cars to see if there is any interconnection between the inertia sensors and the LH supply. I was not aware there was an actual impact related fuel cut on the 928. This is true in many cars, but I don't recall observing that in the schematics before.

I personally would have no reservation using a solid state relay for a fuel pump. It would just need to be correctly speced for the constant current and polarity of the trigger and load.

In fact, I am testing both an SSR and FPDM for my builds, as I need large amounts of fuel for WOT, but normal amounts for cruise and idle conditions. The pump is mapped to a table that has a slight overhead versus the dutycyle of the injectors referenced against the fuel pressure as a safety measure.

Jfrahm

I believe it's clear that it's meaningful to upgrade the fuel punp relay and it seems you do as well if you carry spares and jumper wires. Maybe it's not a big deal to swap a relay on the side of the road but people get killed and injured on the roadside all the time and there are lots of places where having the fuel pump quit might not even afford the luxury of getting to the side of the road.

At minimum I suggest that the fuel pump relay should be relatively new and/or have it's contacts squared up and dressed if it's one of the OEM type I've examined. Better might be a relay made to a higher standard of quality. Modern replacements are out there at modest prices and one could ohm the coil to see if the resistor value might be different. Bosch 0332019103 is a suggested replacement and only $10, it is however a 30A relay vs 40A for the #53. Following an earlier link I see GEBE 993041 (on the linked page but not the same as the one linked) is a 40A. Cheap enough for potentially destructive examination. 30A should be OK but 40A probably has more contact area and the GEBE relays are under $7 shipped from an ebay storefront for the supplier suggested earlier.

I'll grab a few. -Joel.

Jfrahm

--double post--

FredR

Joel,

There are a few things that can cause sudden total failure and the most likely ones are LH failure, rupture of the in-tank fuel pump hose [leading to fragments jamming the main pump] and chucking a water pump impeller [rapid overheating]. Relay failure is possible but statistically speaking with a much lower probability of failure. I have seen questions asked about the quality of Bosch and even Porsche supplied new relays- whether there is any substance in such remains to be seen. Carrying a spare relay is a no brainer givne the cost implication.

The most "challenging" failure i have experienced was on a very busy three lane motorway at night with an ambient temperature of over 40C - dipstick here forgot to install the battery tie down bar and the battery welded itself to the battery box lid. There was no warning- everything died instantaneously whilst i was in the fast lane doing about 90 mph with my wife and [at the] two young daughters. No engine, no lights, no indicators- absolutely nothing. Fortunately i managed to force my way across the traffic and into the hard shoulder. The only positive outcome was that i had fitted a new battery a few days before and took it back to shop advising them that there was "a problem" with the thing and they duly gave me a replacement under warranty . Never told a single lie! Getting the lid open was fun!!!

Jfrahm

I received the GEBE 993041 relays in good order and popped the cover off one to look inside. They look fine. Actually I suspect they have a design "feature" which allows the movable contact to shift position slightly which might prevent a worn spot from forming early.
Coil resistance is a match. Looks like an adequate part.

Given that my experience is that Porsche-supplied relays from the 80's and early 90's are the least reliable of any automotive relay I've worked with lo these many years I'm calling it good advice to replace the critical ones on the 928 before they fail.

Cheers, -Joel.