Jump to content
Volvospeed Forums

Plug N Play Aw50-42Le Paddle Shifter Kit (M4.3/m4.4) Anyone Interested?


Simply Volvo

Paddle Shifter Kit  

122 members have voted

You do not have permission to vote in this poll, or see the poll results. Please sign in or register to vote in this poll.

Recommended Posts

  • Replies 255
  • Created
  • Last Reply

Been uber busy lately guys. Haven't done too much work except some research on TCU's. There are so many different versions and most have different hardware on the board. This shouldn't matter though since all i need to do is piggyback off the pins. For a while i was thinking the CPU's must have some sort of EEPROM. After doing some research I found out most EEPROM's are burnt into the CPU and cannot be read or replaced. Others can be, but with all of the different versions, this isn't a feasible option.

Hopefully Ill be getting a spare TCU soon so i can plug this sucker in already and start coding / tuning.

I will update in the future.

Link to comment
Share on other sites

Gotta love it when you spend money on them and it's still messed up :P

The reason for it is - they use laser etching, and if you ordered a prototype board they probably used the shitty machine which is never calibrated.

Gotta always go over all the traces with an exacto knife on the shitty ones, it'll save you hours of headaches. Otherwise, you need to invest in a microscope (not even kidding lol)

I'm assuming that its all through hole plated right? ie. you soldered everything on the bottom? Because....your soldering looks like sh*t on top and you say the board works :P Also I hate those headers...while they are good for final boards, they suck dick at prototypes - if they are the ones with the metal seats. I hate them, especially because solder loves to flow through them since they are metal, to the other side, and cause shorts, which might actually be what happened and there is nothing wrong with the boards :P

also...I really really hope you increase the trace size for the final...that looks like 10 mil or smaller...go at least 20 mil and 30 mil for the grounds or whatever you desire, just bigger than whats there, and add in tear drops next time- they are really useful for prototypes

Anyways keep up the good work...looking forward to the final product :)

Link to comment
Share on other sites

Gotta love it when you spend money on them and it's still messed up :P

The reason for it is - they use laser etching, and if you ordered a prototype board they probably used the shitty machine which is never calibrated.

Gotta always go over all the traces with an exacto knife on the shitty ones, it'll save you hours of headaches. Otherwise, you need to invest in a microscope (not even kidding lol)

I'm assuming that its all through hole plated right? ie. you soldered everything on the bottom? Because....your soldering looks like sh*t on top and you say the board works :P Also I hate those headers...while they are good for final boards, they suck dick at prototypes - if they are the ones with the metal seats. I hate them, especially because solder loves to flow through them since they are metal, to the other side, and cause shorts, which might actually be what happened and there is nothing wrong with the boards :P

also...I really really hope you increase the trace size for the final...that looks like 10 mil or smaller...go at least 20 mil and 30 mil for the grounds or whatever you desire, just bigger than whats there, and add in tear drops next time- they are really useful for prototypes

Anyways keep up the good work...looking forward to the final product :)

Its through hole. And uhh i soldered this with a $15 radioshack iron and it was my first time doing SMT. My board works fine according to my bench tests and I've had zero problems with those connectors.

Eventually I will up the trace width as current can get up to 1A.

I plan on cutting the final product down to 2" x 1"

I just need to find a spare tcu and i can install the board.

Link to comment
Share on other sites

Cool project.

Looks like your design and board layout needs some work though.

How are you protecting the power rail? Cars have some of the harshest power environments out there. Looks like a simple reverse polarity condition would at the least smoke your LDO, and probably your processor too. Check out the megasquirt schematics to get an idea of how to implement simple input protection.

I don't see any decoupling caps on your AVR.

The crystal is pretty far from the processor and the load caps are way out there too. You want those traces as short as possible. Come to think of it, do you even need the crystal? The internal 8MHz oscillator is probably fine for your purposes.

Will this be a kit or are you only going to sell assembled boards? If you're going kit, I would strongly suggest you go 100% through hole. If it's only fully assembled, you could save a lot of board space by going all surface mount.

How are you flashing the AVR? An ISP header on board would probably be a good idea.

Whoever suggested a four layer board for this is out of their god damn mind. This looks easily routable on a single layer. As it is, cheap board houses charge the same for two layers, so fill the back with a ground plane and any traces you can't run on top. Via stitch down to polygons to spread heat.

Bump the trace width. It's way too narrow. You should be able to do 16 mil traces with plenty of spacing and have room left to spare.

Good luck!

Link to comment
Share on other sites

Cool project.

Looks like your design and board layout needs some work though.

How are you protecting the power rail? Cars have some of the harshest power environments out there. Looks like a simple reverse polarity condition would at the least smoke your LDO, and probably your processor too. Check out the megasquirt schematics to get an idea of how to implement simple input protection.

I don't see any decoupling caps on your AVR.

The crystal is pretty far from the processor and the load caps are way out there too. You want those traces as short as possible. Come to think of it, do you even need the crystal? The internal 8MHz oscillator is probably fine for your purposes.

Will this be a kit or are you only going to sell assembled boards? If you're going kit, I would strongly suggest you go 100% through hole. If it's only fully assembled, you could save a lot of board space by going all surface mount.

How are you flashing the AVR? An ISP header on board would probably be a good idea.

Whoever suggested a four layer board for this is out of their god damn mind. This looks easily routable on a single layer. As it is, cheap board houses charge the same for two layers, so fill the back with a ground plane and any traces you can't run on top. Via stitch down to polygons to spread heat.

Bump the trace width. It's way too narrow. You should be able to do 16 mil traces with plenty of spacing and have room left to spare.

Good luck!

Thanks for all of the suggestions!

Right now the power rail has to protection so I will have to take that into consideration.

The first two caps are for the power supply, the 3rd cap is a decoupling cap for the AVR.

Like I said earlier, this board is a proto layout so I made things far apart on purpose so i could debug stuff(already had to track down a trace the manufacture messed up). Eventually everything will snugged together. You do have a point about using the internal clock. I will look into that.

As of now, I dont really know weather it will be kit or assembled. The high side switches Im using only come in Surface Mount and I am very satisfied with them So i dont want to change hardware.

The back will be filled with a ground plain, I didnt do it in the proto because i wanted to see traces clearly.

Trace width will be bumped up as i mentioned earlier. Some circuits may be pulling 500MAH to 1A, so this is a must.

Thank you for the kind words and suggestions!

Im hoping my TCU box will come saturday so i can start playing!

Right Now im actually considering ditching the onboard 5v Reg(LDO) and tapping off of the TCU's 5V reg, this will probably be more safe as I can draw more current, dissipate the heat easier, use less components and space and also the TCU should have its own protection circuitry built in.

Link to comment
Share on other sites

Sounds good.

The cap you've got there now isn't doing much of anything to decouple the AVR. You want the decoupling cap as close to the power and ground pins of the device as possible, and don't use electrolytics. They're too slow. Get a ceramic cap instead.

I assume you're using integrated high side drivers. You could just use P-mos fets, but you'll need some additional circuitry to drive the gate. It's a balance between cost and component count.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...