• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need a 12v Surge protector for ECU

Thread starter #1
I have a Holley HP EFI ECU that I destroyed when jump starting the vehicle with a high voltage spike. This is what I am being told by Holley Tech. I will have to buy another one but want to prevent that from happening again.

Any ideas ? I looked on the internet and did not find a surge protector for 40 amps DC (fuse size).

I have a 1 farad capacitor. Can that be used to do the job or maybe install another very small battery, like a Motorcycle gel cell sealed battery, wired between the vehicle battery and the ECU to absorb the spike ?

I would rather buy a device than make one but beggers can't be choosers.

Thanks
 

MikeMl

Well-Known Member
Most Helpful Member
#2
Jump starting doesn't cause high voltage spikes in the vehicle being jumped; only in the vehicle from which you are jumping, and only when the engine is running fast at the instant you break the connection in one of the jumper cables. Google "Alternator Load Dump".

I'd say that the design of the ECU is suspect...jumping usually will not harm any automotive electronics, unless the polarity between the batteries is reversed.

Was the engine on the vehicle from which you jumped running fast at the time you connected or disconnected the cables? Best practice is to stop the engine of the jumping vehicle while connecting/disconnecting the jumper cables, or at least slow it to minimum idle.

...
I have a 1 farad capacitor. Can that be used to do the job or maybe install another very small battery, like a Motorcycle gel cell sealed battery, wired between the vehicle battery and the ECU to absorb the spike ?
...
What is the peak load current of the ECU? I looked up its web page. It appears to me that Holley is to aftermarket car electronics what oxygen free speaker wire is to audiophiles...
 
Last edited:

alec_t

Well-Known Member
Most Helpful Member
#3
That linked web page includes an interesting statement :- "This will protect the ECU from any flyback voltage when the transbrake is released."
I would have expected any well-designed ECU to be already protected from such a voltage :eek:.
 
Thread starter #4
I'd say that the design of the ECU is suspect...jumping usually will not harm any automotive electronics, unless the polarity between the batteries is reversed.

Was the engine on the vehicle from which you jumped running fast at the time you connected or disconnected the cables? Best practice is to stop the engine of the jumping vehicle while connecting/disconnecting the jumper cables, or at least slow it to minimum idle.
I was using another standalone battery to jump it. The vehicle's battery by itsef would only click the solenoid. I sent the ECU back to Holley and they could not communicate with it so they deemed it DOA. And yes it just went out of warranty. Original cost was $840, now it's around $1100. ARGH !!!

I cannot repair the unit as they think they have such awesome proprietary circuitry they felt the need to encapsulate it 100% in epoxy. I had hoped that the epoxy was shallow in the aluminum "can" that the board resides in, but noooooooo. see pics. I do not know the peak load current, only the fuse size recommended value is 40 amps.

That linked web page includes an interesting statement :- "This will protect the ECU from any flyback voltage when the transbrake is released."
I would have expected any well-designed ECU to be already protected from such a voltage :eek:.
For the price I would absolutely assume they would put protection in their ECU's. I have a Megasquirt 2 ready to go but the truck is setup for the holley and I think I can scrounge up enough money to get a new Holley ECU but this cannot happen again. It must be all about the money with Holley - Designed-in flaws ? ....Hmmmmm
 

Attachments

KeepItSimpleStupid

Well-Known Member
Most Helpful Member
#5
First, I'd like to refer you to these articles:

http://www.littelfuse.com/~/media/e...utomotive_tvs_diodes_application_note.pdf.pdf

http://m.littelfuse.com/~/media/ele...n_automotive_environment_application_note.pdf

http://www.littelfuse.com/~/media/e...vs_diode_sld_new_product_introduction.pdf.pdf

Next, I need to emphasize the grounding aspect of the ECU connections. It should be a low Z path and this one probably should be where all of the other references terminate. Not just anywhere.

Loose battery cables seem to be the most problem for me. When your car gets serviced or painted, they neglect to tighten the cables.

Before you can the ECU, do one thing: Take it out of the vehicle and short the power connections to ground overnight. The ECU MIGHT recover.
 
Thread starter #7
First, I'd like to refer you to these articles:
Very interesting reads. To add to the install information of the Holley system into the truck, I went out of my way to make this install last a long time by soldering every connection possible, following thei directions to a "T" and highlighting every completed connection on a plotted wiring schematic to avoid start / stop intall mis-wiring. I even bought a dual post optima battery so the starter and alternator charge wires went to the top posts while the ECU related power and grounds went to the side posts in hope of less power interferences. I took over a week to install it while on vacation. It was very aggravating for the ECU to fail after all the install trouble, thus the tendency to stay with Holley ECU - Plug-n-play.

After reeading many forums, it was evident that Holley epoxy encapsulations were considered a big negative issue with users and many went to other ECU's that are repairable. I am still teetering on going back to Megasquirt or try again with holley with battery power mods for a reliable / robust solution.
 
Thread starter #8
BTW - The truck ran flawlessly until I got a job where they gave me a truck to drive and didn't use mine for a year and I disconnected the battery charger / maintainer to use the cord for something else then forgot to plug it back in - the battery was never the same after that then I needed to jump start it one day.....
 

Diver300

Well-Known Member
#9
In my limited experience, potting like that is to hide something, or hide the fact that there's not much there.

I looked at adaptors for LED trailer lights. Adaptors like that are often needed because car manufacturers design around incandescent trailer lights, and LED lights will illuminate at less current, take less current when running and have a much larger voltage at very low currents. Any of those can upset a car that it trying to detect incandescent lights.

One adaptor was some resistors that took similar current to incandescent lights, plus a delayed relay to delay the turn on of the trailer lights. I think they charged about $200 for that. Also, the adaptor isn't legal to use in most countries as with it installed there was no way to detect whether the trailer indicators were working. The adaptor claimed that LEDs run cool, use less power and turn on instantly as reasons why you should use their adaptor, while overlooking the fact that the adaptor could be too hot to touch, used as much power as incandescent lights, and delayed the turn-on by more time that it takes for incandescent lights to turn on. It was of course fully potted and riveted, not screwed, so that buyers couldn't easily find how little it does.
 
Thread starter #10
OK I'm ready to give it another try with Holley HP ECU since it is plug-n-play. Anyone have a idea for me to protect this ECU from voltage spikes ? Capacitors, ferrite beads, a home-made circuit, anything that can be made or bought ?

I need you guy's knowledge / experience.

FYI - the engine was NOT running when it took out the ECU

Thanks !
 

KeepItSimpleStupid

Well-Known Member
Most Helpful Member
#11
1. The TVS diodes, I think will be your best bet. You can get bidirectional and Unidirectional. Use the Unidirectional as they will act as a negative clamp too.

2. it looks as if you did the grounding right or almost right. I had a car once that did have a battery cable and it also had a large ring terminal with a wingnut for the return for everything else. Make sure it's consistent with the rest of the car.

3. You failed to take me up on my seemingly absurd suggestion of shorting the power terminals and ground overnight. I've fixed so many things this way including: 1) A dash car clock (quit after a jump); 2) a bicycle computer (slid along a rug) [someone else's'] 3) An HP calculator that was using commas as the decimal point and countless other stuff. There is physics as to why it works and there is no harm in trying.
 
Thread starter #12
KISS,

I removed the ECU for over a week and had tried it again, this time out of the truck to no avail. I ASSUMED this would have worked the same as draining the ECU via shorting the incoming power terminals together. Nonetheless, I had sent it in to Holley EFI Techs and they could not get it to communicate with their equipment either and deemed it DOA so I thought all bases were covered. Since then, I cut the "Can" off the ECU innards to find it all epoxied in and feel it is destroyed.

I will have to google the TVS Diode and get familiar with it. The recommended fuse size in 40 amps so this too will need to be considered when I pick out a TVS diode.

I appreciate your input !!!!!!!! Let me know if you or anyone else has additional ways to protect this overpriced ECU.
 

alec_t

Well-Known Member
Most Helpful Member
#13

KeepItSimpleStupid

Well-Known Member
Most Helpful Member
#14
I removed the ECU for over a week and had tried it again, this time out of the truck to no avail. I ASSUMED this would have worked the same as draining the ECU via shorting the incoming power terminals together.
It can work, but shorting the power and ground of the device works better. Surges can move charge within an IC and only leakage paths can remove that charge. So, the technique really affects CMOS type of chips. In your case, likely whatever CPU the thing uses. So, shorting power reduces the time constant to drain the charge. In many of the cases, only <30s is required. A computer wireless mouse took overnight. The failure is non-destructive in the sense that it's reversible.

There are no guarantees, but it's worth a shot.
 
Thread starter #16
It can work, but shorting the power and ground of the device works better. Surges can move charge within an IC and only leakage paths can remove that charge. So, the technique really affects CMOS type of chips. In your case, likely whatever CPU the thing uses. So, shorting power reduces the time constant to drain the charge. In many of the cases, only <30s is required. A computer wireless mouse took overnight. The failure is non-destructive in the sense that it's reversible.
OK That makes sense. I'll try that next time, hopefully there isn't a next time.....
 
#17
...I went out of my way to make this install last a long time by soldering every connection possible...
This is actually a VERY bad idea! This makes wires and their connections much more prone to breakage from vibrations and other movement. It's one of the first things technicians learn in racecar and aerospace fabrication.
 

Latest threads

EE World Online Articles

Loading

 
Top