rusefi.com

Well, went back and increased trace width appropriately. At least according to calculations it should be able to handle the max current plus some without putting off too much heat. I had to play around with pin-assignments for hours and hours, but this layout is 1000 times cleaner. Any suggestions a...

I would go wider if you can - remember those traces on the board are VERY short so they can't make very much heat/resistance. What trace width would you recommend? Using a trace calculator, assuming the max trace length of 5inches on 1oz copper fr4 with a 20C temperature rise, 2mm is good for aroun...

Analog/digital inputs can stay small traces - they don't pass significant current. Injector/relay/whatever outputs do however need bigger than 12 mil. Route highest current first, then noise sensitive (trigger, mostly), then everything else. would rotating the Proteus board help at all with routing...

Very very crude routing. All 12mil traces at the moment just to get an idea of trace routing. I'm not a huge fan of how I routed this. Any suggestions for improvement (aside from increasing trace width obviously)? Obviously I do have ground fill, just hidden for viewing. https://i.ibb.co/CwcVBwG/mus...

mck1117 wrote: ↑

Tue Jan 05, 2021 10:58 pm

You mean an ERC failure complaining that the ground pin has no power supply? You can suppress that with a "power flag" (search that in the component picker) and wire to to the GND net.

You're a genius! Thanks!

Oh, and if anyone knows how to fix the GND pin has no source error, please let me know! I looked on a few forums and tried adding a +12V power symbol to a few pins, adding a pspice voltage symbol both with and out power and ground symbols with even more errors. I know it's probably something simple ...

Definitely working way slower than I usually do, but I'm starting to get the hang of KiCad! Mind you that this is about as simple of a design as you can get :lol: Here's the progress: https://i.ibb.co/28vLVPB/mustang-Pn-Pblank2.png Here's the schematic (don't worry, netlist is all connected, just hi...

The pin numbers are reversed, so 1 is 60 etc., but Andrey is correct, you can't invert the connector, I made my own library with the correct pin numbers if anyone is interested.

Edit: here's a picture showing wrong board

Kicad does that too - it just does it only when you push the button telling it to do so. Push the button with the pen on paper. That will assign designators. image.png :lol: The tutorial I was looking at must have been pretty old then because that button doesn't exist in the video. Thanks for savin...

link and name manually? huh? Fusion 360 and Eagle auto assign designators (J1, U2, etc.) to components, and it also automatically syncs the footprints with the schematic symbol. You also don't need to generate the netlist, and the design rule checker is a bit more complete. Again, I'm probably just...

Well, in case anyone was interested, I've been playing around with KiCad, and so far, I'm not really liking it too much, but it also isn't as bad of a transition as I thought. Working with libraries is a pain, and having to link and name everything manually is a pain. I also caught a problem referen...

Looks like the cheapest alternative to headers is something like this: https://www.mcmaster.com/metals/copper-brass-and-bronze/shape~rod-and-disc/diameter~1-16/ I'm going to be using Delrin to insulate the pins as well as be a nice spacer between boards: https://www.mcmaster.com/catalog/126/3841 It ...

mk e wrote: ↑

Sun Jan 03, 2021 7:19 pm

I misread I guess, I thought you were still on about a custom proteus but with an adapter. Sorry....off the shelf+adapter is a fine plan

All good! On the same page now! Hopefully I can get this mapped out in a day or two!

Winner winner chicken dinner Nothing at all wrong with that plan. When you look at the higher end ecus that's how they all do PnP, standard ecu with an adapter of some kind so everyone is using the same HW, it's the only way to support this kind of stuff, you know you have proven HW on every instal...

mck1117 wrote: ↑

Sun Jan 03, 2021 6:58 pm

Nobody knows anything about it. Apparently it was developed in secret which leaves a bad taste in my mouth.

Strange. I share your sentiment since this is supposed to be an open source project, but I'm sure there were reasons. For now I'll stay neutral since I'm new

AndreyB wrote: ↑

Sun Jan 03, 2021 6:12 pm

We are 3-8 weeks away from https://rusefi.com/forum/viewtopic.php?f=4&t=1913 potentially changing the hardware reality of rusEFI.

Is there any additional info on this? Should I just wait to use this platform?

mk e wrote: ↑

Sun Jan 03, 2021 6:18 pm

That is by far the best solution. You won't ever regret having spare I/O or having a setup that is fully supported.....and its still 1/2 the price of the less capable MS PnP so pretty good there too.

I should be specific, I still will be doing an adapter board, but using proteus.

Or maybe just uprate the 4 current injector drivers and leave everything else? semi-sequential and waste spark is a fine combination and low cost. Haven't seen an update in a while but there was a member who did a PnP board for a 5.0 as well using the mre which is completely viable option, but if I...

rusEFI strategy is discussed on slack. best way to help rusEFI as of Jan 3, 202 is NOT by designing any new boards. Heard loud and clear! I hope you know that I'm just brainstorming and not trying to subvert anything. I didn't know there was a slack channel. Is that something that would be helpful ...

if you were really desperate you could cut lengths of 1/16" brass brazing rod Not the worst idea for sure! I'd like to pick your brain about mk e's idea of taking the proteus microcontroller and adapting it to the mre board layout and connector. It would be cheap, and it would allow regular he...

Adding a shit d-sub is cheap, but everything adds up....you have way too much on your your list to hit the price point. You are talking about a mre, not a proteus at your price target. You can't afford adapter boards, knock, ETB, you don't need any of it anyway so kill all that and make the board a...

One thing we couldn't find when Gwendal was designing that NB2 PnP board was usable header pins to go between the Proteus and adapter board. They're 4mm pitch, but pretty large diameter. I think on his he just cut the plastic shell off of some normal Ampseal connectors and used the as adapter heade...

Meh, pretty much everything on an ECU is pretty tolerant of that mess, plus you're in a nice cast aluminum faraday cage! Yeah... I just took an EMI design class so there's a lot of "what-ifs" that could happen. Though I will say that the "cage" might actually act more as an ante...

I suggested that 4 pages ago : :lol: Well, as promised, I am humble enough to eat my words! I'm not sure how I missed that! I must have thought you were talking about the dev boards where people wire the boards with physical wires to a breakout board. So yes, apologies are in order. At LAST!!!! we ...

there is another option, and that's to make an adapter board between proteus and your car like this: https://rusefi.com/forum/viewtopic.php?f=4&t=1701 Aaaand that's the best idea I've heard this whole time! :lol: Never even occurred to me! :oops: There's plenty of room inside the stock case for...

Backwards, you see it most at low rpm, low power when injector pulses are shortest so error as a % is highest. The ignition options you won't see at any rpm. Good to know. Either way It'll be setup for just about any option. I will be the first to admit that I am not sharpest tool in the shed most ...

does it have a missing tooth? Yes. The stock TFI also gives the options of sequential injection and distributor, waste, direct ignition. And the crank/cam setup will also support distributor ignition. The difference is more accurate spark timing during acceleration. You are correct. Though the adva...

You have or have access to until you graduate? ....and you still need to buy all the parts hoping to sell the units. At $300 they will surely sell if that is the going rate for a replacement used ECU but you'll need to require a core to steal the case from and connector? so a lot of labor and shipp...

who is going to fabricate 10 units for this to help 10 people? what kind of investment would be needed to fabricate 10 units? Well, I'm not sure how this open source platform license works exactly. I feel it would be both immoral and rude to not give credit where it's due and mck1117 has put in qui...

You mentioned a 4 tooth crank sensor I think it was? That is of no value, its the same accuracy/data you already have on the distributor trigger. Sorry for the confusion. The crank is a 32 tooth not 4. As far as removing the distributor and adding a cam sensor, after you add a proper crank position...