Modifying an OBDLINK SX USB Adapter for HS/MS-CAN relay managed by RTS

[ham_op]

@ForScan

In my 2003 Ford Escape 3.0L V6 car, with a laptop hooked up to the OBDlink SX (OBD-II port connector & USB cable are tiewrapped to prevent potential disturbance by feet on control pedals) running Win 7 Home Premium (yech McNazi$oft crap OS), some condition is crashing ForScan program, generally, but not always, when I'm running OSCILLOSCOPE and it seems that the problem appears to be a buffer over-run behavior, which is consistent with a buffer that is too small for the amount of data being streamed. I originally thought that the problem might be caused by the Ford side of the bus over-running the adapter, but now I've determined that its on the adapter side of the bus, it looks to be a hand-shake problem. However, I notice that only 12 or 13 items may be displayed, while the TABLE window displays ALL of the assigned monitor points, the OSCILLOSCOPE 'window' does not incorporate a side of window vertical scroll bar, so any monitored item beyond the bottom of the OSCILLOSCOPE screen isn't viewable. From a programmer's perspective, what should be implemented is a DASHBOARD item counter that prevents more monitors than can be displayed in the OSCILLOSCOPE window or provide an OSCILLOSCOPE scroll-bar to enable scrolling the traces up to view the below window traces. Although there does seem to be some correllation between the over assigned monitors on the OSCILLOSCOPE page and program crashes, the TABLE window when in focus also crashes the program!

Via the DASHBOARD I deassigned enough monitor points to only display 12 traces and since doing so the program has not crashed. Of course this may be simply coincidence. Only further extended operation will suggest the truth.

[someone972]

Hmm, I suppose I could put an oscilloscope on the RTS line and see if it changes state at any time during regular use. Not sure when I'll get around to that though.

[someone972]

Took a closer look at the board today. Was able to confirm that the 5 pin unpopulated header next to the OBD port is in fact a PIC programming port for the PIC24HJ128GP502 main chip. The pin that the RTS line is hooked up to is the PGEC2 pin of the chip, so it's likely that the firmware flashing process is using the RTS signal as the data clock for the chip. It shouldn't cause any harm during normal operation to assert that pin low I don't think.

I was able to get my circuit rigged up with a DG9415DQ dual 2:1 analog switch chip. After a couple of screw ups on my part I was able to get it to work with FORscan's auto switching! Now I need to do a permanent install; breadboards hanging off of the OBD port isn't the most stable . I will hopefully be able to do a write up of sorts on how to modify the red OBDLink SX like I have it once the permanent install is done.

[Microbe]

Took a closer look at the board today. Was able to confirm that the 5 pin unpopulated header next to the OBD port is in fact a PIC programming port for the PIC24HJ128GP502 main chip. The pin that the RTS line is hooked up to is the PGEC2 pin of the chip, so it's likely that the firmware flashing process is using the RTS signal as the data clock for the chip. It shouldn't cause any harm during normal operation to assert that pin low I don't think.

I was able to get my circuit rigged up with a DG9415DQ dual 2:1 analog switch chip. After a couple of screw ups on my part I was able to get it to work with FORscan's auto switching! Now I need to do a permanent install; breadboards hanging off of the OBD port isn't the most stable . I will hopefully be able to do a write up of sorts on how to modify the red OBDLink SX like I have it once the permanent install is done.

Hey there someone, did you manage to get the auto ms/hs working on the new red SX ? Thanks ! interested to know.

[dadams2169]

Took a closer look at the board today. Was able to confirm that the 5 pin unpopulated header next to the OBD port is in fact a PIC programming port for the PIC24HJ128GP502 main chip. The pin that the RTS line is hooked up to is the PGEC2 pin of the chip, so it's likely that the firmware flashing process is using the RTS signal as the data clock for the chip. It shouldn't cause any harm during normal operation to assert that pin low I don't think.

I was able to get my circuit rigged up with a DG9415DQ dual 2:1 analog switch chip. After a couple of screw ups on my part I was able to get it to work with FORscan's auto switching! Now I need to do a permanent install; breadboards hanging off of the OBD port isn't the most stable . I will hopefully be able to do a write up of sorts on how to modify the red OBDLink SX like I have it once the permanent install is done.

someone972 im curious too did you get it all up and working and have any pics yet as i too have the red one and am ready to do this too.

[fxnick]

Went to do the mod to my adapter, but the inside is totally different.
Here are some pics of the red rev. J board it it helps anyone.

2019-11-08 20-21-26-1.jpg (493.35 KiB) Viewed 15579 times

[cmpink]

Took a closer look at the board today. Was able to confirm that the 5 pin unpopulated header next to the OBD port is in fact a PIC programming port for the PIC24HJ128GP502 main chip. The pin that the RTS line is hooked up to is the PGEC2 pin of the chip, so it's likely that the firmware flashing process is using the RTS signal as the data clock for the chip. It shouldn't cause any harm during normal operation to assert that pin low I don't think.

I was able to get my circuit rigged up with a DG9415DQ dual 2:1 analog switch chip. After a couple of screw ups on my part I was able to get it to work with FORscan's auto switching! Now I need to do a permanent install; breadboards hanging off of the OBD port isn't the most stable . I will hopefully be able to do a write up of sorts on how to modify the red OBDLink SX like I have it once the permanent install is done.

Did you ever get the write up done for this? I would be interested in seeing how you did it before I try it my self.

[bluto1]

Went to do the mod to my adapter, but the inside is totally different.
Here are some pics of the red rev. J board it it helps anyone.

I have a rev. J board also and just started modding it. I pulled up the data sheet for the FTDI FT230XQ. Pin 16 is the RTS pin. The trace is very small but it seems like I could solder a very fine wire to it. In the second picture, I inserted a small wire through the hole near the PIC chip. I think I will try scratching a little bit of the coating off of the via hole on the back side and soldering there.

[cmpink]

Went to do the mod to my adapter, but the inside is totally different.
Here are some pics of the red rev. J board it it helps anyone.

I have a rev. J board also and just started modding it. I pulled up the data sheet for the FTDI FT230XQ. Pin 16 is the RTS pin. The trace is very small but it seems like I could solder a very fine wire to it. In the second picture, I inserted a small wire through the hole near the PIC chip. I think I will try scratching a little bit of the coating off of the via hole on the back side and soldering there.

I had come to the same conclusion. Just waiting on some parts and looking for my small solder tip.

[dadams2169]

has anyone got the red board mod working yet?