For discussion of Speeduino compatible boards designed / built by other members of the forum and for guidance around making such a board
User avatar
By pazi88
#37062
Raku wrote: Mon Aug 12, 2019 4:08 pm best coil drivers to use is probably bosch bip373 driver.they have overcurrent and overtemp protections. unfortunately expensive and quite hard to comeby .
Those are quite oldschool and not even IGBT's. Not fan of those. But can be used too of course.
By Rafa_bmx
#38088
exvils wrote: Mon Jul 22, 2019 5:25 pm So, on Friday started hillclimb race day, car was on Friday on stock ecu (with MAF/cps/lambda) but did not had power and other racers told my driver that it sounded like it was on 4cyls. On Saturday there was qualification day, at 1st round car drove 4.04min lap with stock ecu, driver than swaped stock ecu for speeduino (disconnected maf/cps/lambdas) and car drove 1st 500meter of race like it was back with full power, then lost power (coils) for rest of track, still went 4 sec faster. . I came to race after this and started investigating what happened. Coils on cylinder 1 and 6 melted, also i think cyl nr.2 (and maybe 5) wasnt working (but coils were allright), so i opened case and found this :lol:
Any ideas what happened? Coil dwell was setup at 2.0ms. It was working for past 3 weeks, even broke LSD 2days before race while tunning at local airport. Tho i dont have any logs from race when it broke.
Image
Image

it happened the same to 2 of mine, exctly the same! i think bmw coils pull a lot of current
User avatar
By pazi88
#38094
Really strange that those things happen. Those IGBT's don't even get warm to touch in normal operation. I have seen even 600hp turbo e30 doing burnouts flooring the throttle for 5 minutes straight with this PCB design. And no problems what so ever. You sure you use genuine components and not eBay stuff?

At least in that picture the it's strange that only one IGBT has got warm and others are fine. If that would be caused by coils taking more current than PCB can handle, all traces should be melted. Also if that would be caused by for example ignition output stucking on, then there should be 2 burnt IGBTs. But one like that can only be caused by that one IGBT grounding permanently. At least in that pic it looks like the original 25 year old insulating sheet is used. Which can be easily damaged when doing this kind of thing because it's old. And doesn't even cover the whole area for the IGBTs. It could have had damaged during installation so that the IGBT grounds the coil permanently through the insulation.
User avatar
By exvils
#38796
as i said, it broke on oem ecu (car still had original coils from factory while seeing heavy use) and swaping oem ecu for speeduino (for fixing the problem) when you know something is wrong with 2 coils not firing was a bad call from the driver. After i fixed those blown traces speedy still works (half year later), so thx for this custom board pazi88
I would be much happier if there were fuses on coils on that car since that accident ( :D my level of trust for old, non-smart coils went lower, tho i still trust m54 coils :) )
User avatar
By PSIG
#38797
exvils wrote: Sun Oct 27, 2019 2:07 pm... my level of trust for old, non-smart coils went lower, tho i still trust m54 coils :) )
My view is different, and anything that can pull enough current to cause damage should be protected by fusing or other methods, 'smart' or not. Damaged wiring and connectors can be much more involved and expensive than repairing a PCB trace, so I'd protect whatever needs protecting and sleep better. ;)

BTW, keep in-mind that automotive fuses are designed to never blow at their rated amperage :shock:, e.g., a 10A fuse will not blow at 10A. At 11A it may take many minutes to blow, so be sure to size your fuses to the load, or to blow below the wiring's calculated max-rated load. Fuses and similar stuff is easy to add to existing wiring.
8-)
David
User avatar
By pazi88
#38798
I have been doing some investigation on that burnt traces problem and can't seem to reproduce that no matter what. Like 5000rpm, 8ms dwell and half an hour running on the bench and the ignition side isn't even warm. One thing that I haven't tested yet is what happens if the actual coil(s) are shorted. Still, I have designed next revision with thicker IGBT traces and better clearance to ground plane. But not sure if that will help anything, because I can't cause any problems with current design in normal operation.

I have seen current limiting components on megasquirt based clones after the igbt's
But problem in those is that the components need to be rated to 10 amps at least to provide sufficient spark in high power turbo use. But on the other hand if something goes wrong and you get 10amps continiously, it can still easily lead to burnt something. And they normally use something like 2amp current limiting because of that. And that leads to very weak spark.

OEM's seem to incorporate active current sensing by the processor and prevent this kind thing in ecu code, but not sure how to do that for speeduino.
By theonewithin
#38800
Speeduino I wouldn't use to do monitoring of current etc. It has enough to do already.

An addon board (BCM style) may be a better idea that is in control of ignition relays probably best idea.

Could have it monitoring any circuit you like.
  • 1
  • 12
  • 13
  • 14
  • 15
  • 16
  • 35

Hi, I am trying to assign Signed values to the x-a[…]

Vr Conditioner Noise when cranking

New version 202305 don't fix the issue. Now after[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Hello, I bring news!! Let me tell you that after […]

Still can't find what you're looking for?