No serial connection

Jeffs_Alpine · Post by **Jeffs_Alpine** » Wed Jun 02, 2010 9:34 pm

I originally purchased my V4 (I'm assuming this based on the history images) MegaJolt last year. I assembled it at that time.

I have just finished installing the car's motor (67 Sunbeam Alpine with a few modifications) and wanted to get that running well before introducing a new variable - the MegaJolt.

So with everything humming, I attempted last night to connect with my laptop. First tried the Firmware uploader. Wouldn't connect. I've searched this board and found all of the other occurances of this issue and found some helpful tips. Unfortunately none solved my problem.

I've taken the controller out of the car and set it up on my office bench. I've tried connecting with three different computers, stripped down the cable and re-wired it to use only the 3 wires outlined in the Serial PDF document, I found a wicked cool serial port monitoring program, which shows me the attempts at communicating with the box, on a valid port - but no connection. It always hands with 'ACK...'. Very frustrating.

I know the 5volt regulator is working, since the green led is on. f I had an oscilloscope I'd check the cpu's clock, but the rs232 chip doesn't appear to require any clocking, so it shouldn't matter.

So I wanted to ask a couple of questions (to BrentP):

- if the cpu was toast, for some reason, would the serial comm protocol get past the ACK... stage?
- do you have the ability for me to send in the finished board and for someone to swap it to determine whether it's dead or something else?
- has anyone ever seen a cold soldering joint cause the board to go deaf?
- and any other thoughts, suggestions anyone might have?

I've spent a day on various connectors, cables, and cpu's, but nothing. I've set the port driver to 34800 and still nothing. I'm really looking for any new ideas.

Thanks - Jeff

Post by **brentp** » Wed Jun 02, 2010 9:48 pm

Hi Jeff,

An alternate way to see if the CPU is "alive" is to wire it in to an engine already running the EDIS system. If you can verify it's controlling ignition advance through the use of a timing light (advance should change as RPM varies), then the CPU is working, and the problem lies in the part of the circuit that communicates with the PC.

Other than that- could you post some good pictures of your assembled board, top and bottom?

And, could you post the actual output of the firmware uploading routine? Note, this is not normally needed as the chips come pre-loaded with firmware...

Jeffs_Alpine · Post by **Jeffs_Alpine** » Wed Jun 02, 2010 10:33 pm

Brent, thank you so much for the incredibly prompt response.

Here are four (very large) photos: front back, taken from two angles.

Thanks again!

Jeff

Jeffs_Alpine · Post by **Jeffs_Alpine** » Wed Jun 02, 2010 11:36 pm

Hi again Brent,

Re-reading your last post, I wanted to respond to the other two points you've raised.

Using an existing EDIS system is not an option right now - I don't have access to an auto with that type of system.

The actual code that's being output:

when attempting to use v4 of the configurator, I get "Could not read global configuration data from controller: A communications timeout occurred". I can see that the configurator sends a series of "VSVSVS...", I'm assuming, as its handshake attempting to communicate with the device.

when I use a dos-prompt and the s-record uploader, it simply says: "Waiting for HC08 reset ACK...", until I control-C out of the loop.

The serial port monitor that I've been using "Free Serial Port Monitor" - highly recommended - shows IRP_MJ_READ UP FALSE 0x102, DOWN FALSE 0x00. No handshake from the device.

Hope this helps.

Jeff

Post by **brentp** » Thu Jun 03, 2010 2:46 pm

Hi Jeff,

Your linked pictures are not showing up, as they're asking for a passsword to be viewed.

What I would recommend is to shrink them and then attach them to the thread using the attachment feature.

Jeffs_Alpine · Post by **Jeffs_Alpine** » Thu Jun 03, 2010 3:28 pm

Hi Brent,

I have access to just too many servers - I'd already logged into the one that I'd placed the images on, and it worked just fine for me

Sorry - please take another look.

Jeff

Post by **brentp** » Thu Jun 03, 2010 3:43 pm

Hi Jeff,

the solder joints on the back side look a bit marginal. What type of iron are you using?

Jeffs_Alpine · Post by **Jeffs_Alpine** » Thu Jun 03, 2010 3:53 pm

Brent,

I don't recall the make, but I know that I intentionally avoided getting the board too hot, perhaps too much so.

I wanted your perspective before going through the re-heat/add solder process.

Jeff

Post by **brentp** » Thu Jun 03, 2010 5:26 pm

When soldering, you want a good 2-3 seconds of pre-heating using a 15-30W fine tip iron- and use fine soldering wire.

There are some good soldering tutorial videos on youtube- see here for one example:
http://www.youtube.com/watch?v=I_NU2ruzyc4

I would recommend re-touching all of the solder joints and then attempting a re-test.

Let us know what you find out!

Jeffs_Alpine · Post by **Jeffs_Alpine** » Thu Jun 03, 2010 5:39 pm

Hi Brent,

Thanks for the solding tips.

I've just finished re-touching all of the back-side joints. I'm using an older, fine-tipped Weller. I touch the tip to the joint, allow the pad and the existing solder to solidify (1-2 seconds), then add a touch of new solder, allow it to flow (1 second), and then remove the tip. You can see the results on the flip side - the solder flows through the board.

I'm now testing on the second computer - the first one yielded the same results as earlier tests. Argh....

I'll post the final results shortly.

Jeff

Jeffs_Alpine · Post by **Jeffs_Alpine** » Thu Jun 03, 2010 6:26 pm

Brent,

Finished testing on the second computer (WinXP sp4) and still no ACK...

I tried to attached a zip, but it was too large I guess - told me it was empty. So here are images of a screen shot of the desktop showing the dos-prompt (cygwin) window and two new shots of the back of the board. I was tempted to clean the board with some lacquer thinner to remove the flux to make the joints more visible, but, decided against for now.

I'm starting to think that the vacuum advanced Lucas distributor isn't all that bad......

Jeff

Post by **brentp** » Thu Jun 03, 2010 6:53 pm

Jeff, where are you located (part of world)? Perhaps another user can help take a look at it for you.

One side point, did you verify all of the voltages at all the test points documented in the kit build instructions?

Jeffs_Alpine · Post by **Jeffs_Alpine** » Thu Jun 03, 2010 7:25 pm

Brent,

I tested the voltage points after finishing the regulator installation and they were all good. I just rechecked them and they're still good. The V++ on the darlington array seemed a little low, +11.94v, but I'm assuming that's ok.

I am in Sunnyvale, Ca - just south of San Francisco, near Palo Alto.

I'd welcome any suggestions and/or help at this point.

Thanks,

Jeff