Anybody else bricked their MS-II?

[Jim_Blackwood]

I've got 3 of them, out of three controllers we've had 100% replacement now. The last one apparently bricked when I clicked OK on the dialog that warns of signature mismatch. All communications lost. Just wondering if this is a common problem, and if it is, why was that option allowed anyway?

Jim

[jsmcortina]

No it isn't a common problem.

First off, try using "Detect" from Communications -> Settings.
If that doesn't get you anywhere, then reload the firmware. You may need to use the boot-jumper on the MS2 card.

James

[racingmini_mtl]

Have you tried reloading the firmware using the bootloader jumper?

A signature mismatch could lead to corrupted data which could lead to unexpected firmware behaviour. That could have an impact on the ability of the code to continue running normally. I don't see that as a high probability event but it is possible. Using the option should be avoided with firmware versions that are far apart but should not be a problem when upgrading to the next version. If you have any doubts or are not sure, don't use the option.

Having said that, It would be very surprising to have a truly 'bricked' MS2 simply from using the wrong tune file format. If that has happened to you more than once, I would suspect that you have some other hardware issues that would create that and that would be related to the MS2 being connected to the PC. The CPU cannot erase the serial monitor without using an external device so you cannot brick the MS2 by a simple command from the serial port.

Jean

[Jim_Blackwood]

All three MS-II cards were in different controllers and all three controllers are now working fine with new MS-II cards so I'm struggling here to find a common factor. None of the three respond to the firmware installer, all of the replacements did. There is no communication with any of the three, and I just think that is very odd.

Jim

[Matt Cramer]

Did you use the TunerStudio firmware loader or the downloader included in the firmware ZIP file?

[rickb794]

Maybe you have a port conflict or some other thing is happening that is pulling down the 5 volts, that will shut it off.

I'm not sure you can actually brick one, even if you abort a firmware load. You just jumper the bootloader and go.

In the 8 plus years I have been working with Megasquirt I have never seen one "bricked". (three in a row is off the charts)

I bet if you pull it out of the car and power it up with a low power source (wall wart) it will run.

I will gladly run it on my test bench and send it back No Charge if you Mail it to me with return postage.

[panel]

Jim.....where are you located.....don't see it in your profile

[DaveEFI]

All three MS-II cards were in different controllers and all three controllers are now working fine with new MS-II cards so I'm struggling here to find a common factor. None of the three respond to the firmware installer, all of the replacements did. There is no communication with any of the three, and I just think that is very odd.

Jim

Very rare for a processor to fail for no obvious reason, and nothing else failed on the main PCB. My guess would be 12v getting on the 5v rail - or some massive spike on the 12v rail. But even then I'd expect other damage.

Have you kept the failed processors? They can be repaired for less than the cost of a new one. But may give a clue as to what caused it.

[Jim_Blackwood]

I agree it is very strange.
But I am beginning to suspect I could be the ultimate Beta tester, you know, the one who finds every possible flaw purely by accident. Not a good feeling. It makes me wonder if perhaps I should find another interest, but by this point I am too invested in the MS community to easily back out. After all, I've been here since MS-I and the early version mainboard. I've not always been an active member of the community, but I have done things like early EDIS with PWM IAC implementation and Phil's knock detector, so I'm not a total noob.

Rick, if you are serious I think I should accept your offer. It would be very helpful to have someone else who is familiar with the equipment verify my results. Heck I'd be the first to admit that human error could play a part, but the facts remain what they are and replacing the MS-II board in every case resulted in getting the controller working again. So as rare as these problems are, I can't rule it out, even if it appears I might be the only person who has ever had this problem. I agree that would seem statistically highly improbable, put that still leaves it within the realm of possibility. I also had a MegaShift board shipped to me without the required initialization steps being performed. Somehow it slipped through the cracks and communications could not be established. I still have that board on my desk where I've used it to reference component locations, a free replacement was sent after the problem was identified. And I hate that, but there was nothing else that could have caused it. I sorta feel like a defect magnet. Weird right?

My troubleshooting processes have not been absolutely perfectly systematic, so I realize inconsistent results could creep in. But I've tried my best to get consistent process and results. I'll be pulling the controller back out today, removing the one working MS-II card I have and going through the firmware load process one more time for each chip. I will try the TS firmware loader again as well just for more redundancy, and when I finish that up I'll plug in my MS-III upgrade and see if I can get the simulator (JimStim) to output injector pulses and a SAW wave, and if that works I'll see how the car will run on that. Which if it works will render these other processors as spare parts. But at this points I'm really feeling like a few spare processor chips would be an excellent backup plan not just for me, but for my brother who is now running two MS-II controllers.

I've been using the V2.0 (B&G) firmware loader. Also I updated my profile with my location. I'm in Florence, KY just outside the Cincinnati beltway. I have no idea where the closest other MS user is to me, other than my brother 180 miles away, but I do feel pretty isolated and I'd like to change that.

Jim

[jsmcortina]

I've been using the V2.0 (B&G) firmware loader.

This may be the problem. Use the current firmware bundled with MS2/Extra instead or the TunerStudio loader.

James

[Jim_Blackwood]

Using the firmware upgrade tool in Tuner Studio:

I used my current MS-II card as a control, checking connectivity before, during, and after all the other tests. The TS loader does not require you to jumper the two pins on the board, in fact when I did jumper them I got a "No Controller Detected" error (NCD). With the pins not jumpered the card was detected properly, and the firmware upgrade and backup reload proceeds as expected. I vigorously cleaned the socket with acetone and a short bristle acid brush after initial testing, this did not cause any apparent deviation in results.

Each of my three suspect MS-II boards exhibit similar symptoms with one exception: The V1.0 board lights the FP LED when power is applied whereas the other two (V2.0) flash once very briefly upon application of power. The common indications are, all three front panel LEDs are lit, and I get a pulsating light on the IGN LED. All three give an NCD error when the "Test" button is clicked. Incidentally, a MS-I chip tested gave an NCD error but JimStim indications were that it was functioning, as Ign and Inj LEDs were pulsating and FP was lit.

I'm not sure what else I can do to compare these cards, it seems pretty obvious that I have 3 bricked MS-II cards. But I'd be perfectly happy to send them out to someone else to test.

Incidentally, one of the two V2 cards initially was detected but when I removed it and then re-tested it later it was not. So far that has been the only inconsistency in the test and I re-tested/re-inserted that card at least a half dozen times trying to get a repeat but it never happened. Testing of the known good card was consistently good once I figured out that the test had to be run with the pins not jumpered. Unfortunately I did not run "Detect" at that time so I do not know if it would have worked that one time or not.

Anyone feel like taking a crack at these? As unlikely as it seems they all three seem to be bricked. Not a clue why.

Jim

[jsmcortina]

I can't help you with the green MS2 card.

How do the later cards respond when using the MS2/Extra firmware loader?

James

[ashford]

about 8 years ago i had something similar happen. it wouldnt reflash extra code. i then flashed an old b&G code with ms2 downloader then flashed the extra code back onto it

[Jim_Blackwood]

I haven't tried that. I presently have the unit built as a MS-III and am checking the configuration. As you are aware, going back to the MS-II is a bit more work than just swapping cards.

Jim

[jsmcortina]

As you are aware, going back to the MS-II is a bit more work than just swapping cards.

In what way?

James

[jsmcortina]

about 8 years ago i had something similar happen. it wouldnt reflash extra code. i then flashed an old b&G code with ms2 downloader then flashed the extra code back onto it

I've seen one other user post that, however it doesn't make any sense to me.
Both loaders are sending an "erase all flash" command and then programming the flash.

The only possible difference I can see is that the old loader does a verify, so it might alert you if you had serial corruption?

James

[racingmini_mtl]

One other thing that could be happening here with the switch between the B&G code and the MS2/Extra code is that the chip on the MS2 card is identified (and sold) as a 64k flash chip but the MS2/Extra code does use it as if it's a 128k chip (recent versions of the B&G code might also do this but I don't know; earlier versions were limited to 64k). In theory, both chips (64k and 128k) versions use the same die so they both have 128k of flash but it is possible that the extra 64k is either not tested at the factory or does not pass some tests.

I'm not sure exactly how this would explain any of what has been observed in the past nor why it is not seen more frequently. Maybe someone has looked into this in more details.

Jean

[Jim_Blackwood]

All interesting food for thought. But if I can't even confirm the existence of the chip isn't it unlikely it can be flashed, verified or even erased?

James, what I meant was, it's all boxed up so I'll have to break it back down to the bare board to remove the MS-III what with the standoffs, the larger case and all before I can plug the MS-II back in. I can do that, but if it won't even recognize the chip exists either with the B&G 2.0 or with Tuner Studio (latest version I believe) then what's the chances the result will be any different with the other loader? I mean I can try it, but I wouldn't hold out much hope of success.

Jim

[rickb794]

Any chance a battery charger was involved in this "bricking"? The choppy waveform a battery charger presents will do bad things to digital stuff.

[kjones6039]

Any chance a battery charger was involved in this "bricking"?

Funny you should mention that, Rick.

The only time I ever had issues with an MS2 card was when my grandson replaced the alternator in his truck with a POS from the scrap yard. That chip has never been quite right since. A battery charger seems to me to be somewhat analogous to the flakey alternator.

Just a thought.................

Ken