risu wrote:I was thinking about little pcb which consists of two so8 LM1949 and those 0.1 resistors and it would fit to normal DIL8-socket. Normally there's no need to relocate TIP42 which is already on MS-board, and FETs too.
FETs leg on MS-board would have to be disconneted from MS-board and it would be connected to this new PCB:s resistor. Though I'm not sure if it's a different GND in MSv3. At least on 2.2 it's the same ground with driver and fets. It's just a thought to minimize space and components. I admit I haven't searched through the datasheets so there might be something I've missed.
I've no intention to dismiss your effort with current layout, looks like you've done a good job with it.

I didn't think you were dismissing my work. I just didn't get what you ment at first but I got it later after my last post.
Even though it's an interesting idea, there are a few things that could make this impractical. The first thing is that the LM1949 is no longer available in an SO8 package, at least not that I could find. The second aspect is that the LM1949 is designed to drive a single injector. You probably could cheat and drive 2 injectors but you'd have to change the sense resistor (half the resistance and 4 times the power rating) and make sure the injectors are well matched so that they both receive the same current and open at the same time. There might also be a need to change the FETs because I'm not sure what can be driven by the LM1949.
Another point is that the LM1949 requires a non-inverted pulse signal which is not what is provided by the current code versions on MS1 and MS2. There could an inverter on the board or the code could be changed but that last option, while not difficult, would also require a change of the pull-up resistors to pull-down resistors. This is similar to what is done on Microsquirt.
The last point is that the MSV3.0 and V2.2 have differences in their injector driver circuit. That could require 2 different daughter boards depending the PCB version. It would also be nice to be able to use the sense resistors already present on the V3 board.
All these concerns might no be big issues. Actually, I'll have a look in more details at all those to see what would be the best solution in terms of small footprint and minimal connections.
Jean