My Vectrix is now a doorstop (failed firmware update attempt)

I have a good motor controller - dash - ICM - and Charger all ready to go if you need replacements.

Obviously it would mean shipping it from here to you in the US but thats nothing difficult.

Plus they ALREADY have the Lairds firmware fitted and work great.

Basically i had a front end smash due to a myopic Mercedes driver. But apart from bent forks bent steering stem and yoke, all my electronic systems are good to go and still function perfectly.

I've been in touch with Vectrix to get spare parts (just need a steering stem and handle bar clamp) but other than a quick reply back the next day they dont seem to be forth coming so there you go seems abit hard to deal with sometimes.

So..... if you are interested PM me and maybe we can come to some arrangement - fitting of most of this stuff is a piece of cake if you take your time.
The ONLY thing that would need doing once parts are fitted is the Motor would need calibrating to the controller but this is a 2 min operation that can be explained if you get to that point.

Cheers

Chris

Incidentally I know it doesn't help but I managed to kill my dashboard when my battery on my laptop failed in the middle of re-programming (thought it was plugged in but wasn't) and I had to get a replacement dash for my bike.

So i know where you are at with it.

Incidentally I know it doesn't help but I managed to kill my dashboard when my battery on my laptop failed in the middle of re-programming (thought it was plugged in but wasn't) and I had to get a replacement dash for my bike.

So i know where you are at with it.

*gasp!!*

I am not a Vectrix expert, but I am a firmware programmer. What is likely going on is some flash part on the scooter has an incomplete image or has gotten erased. There should be a way to reprogram it under almost any circumstances. Believe me the guys that developed the firmware bricked their cycles 1000's of times. There has to be a way out, but it might take a conversation with a Vectrix engineer to find out all secret incantations. It might, for example require jumping some pins or it might involve soldering on a header you don't have. In the worst case it might involve some expensive hardware and replacing the circuit boards would actually be cheaper. Good luck!

Hello Folks,

I am having some guilty feelings for encouraging owners to 'improve' their bikes. BUT, I plead innocence inasmuch as the majority of difficulties we are having are a direct result of the (mostly) unnecessary complexity of the Vectrix programming.

No doubt some of you will say that 'it's a poor workman that blames the tools' but in this case the 'tools' are faulty.

The problem, that Scott has now got, is down to the fact that when re-programming the motor controller, the process first of all deletes the 'boot file/information' that start the CanBus communication along with all other data ( the files that the diagnostics programme uses for this is the CLR24_X1.hex and the ERA24_X1.hex ) and only then proceeds to re-install the boot files ( Using the PRG24_X1.hex )and the 'new' programme. The CanBus communication continues working through this process because a copy is held in volatile memory for use during the re-programming process.

Obviously, any disruption to the bikes power supply (like switching it off) whilst this process is in motion will result on an inability to restart the CanBus, It had no way to boot up. Vectrix know/knew about this all along but have provide no recovery mechanism other than a 'factory re-install of the 'boot files'.

The install procedure, for the modified software, describes how to deal with any hang-ups that may occur during the Motor Controller programming procedure. It are not foolproof, but if followed, will contain and correct the vast majority of problems, provided of course, that the bike is not switched off.

In Scott's case one of two things have happened. Either, some hitherto unknown programme lock-up has fouled the process in such a way as to be irrecoverable. This has not happened before to my knowledge.
Or, when something went wrong Scott failed to follow the recovery procedure to the letter.

I am not looking to lay blame or to make excuses. I seek only to describe what has happened as far as can be established.

We are all suffering from a serious dose of overcomplexity generated by the designers of bike's control gear.

The cure to Scott's problem is, more than likely, a Factory reinstall of the boot files or a replacement controller with the files already fitted and intact.

To my knowledge, this is only the second time that this has happened. In both cases the diagnostics programme asked for the file CLR_X1.hex. This 'request' by the diagnostics programme is a clear indication that the diagnostic programme itself is not correctly installed on the computer being used. In the first case, around last Christmas time, the diagnostics programme did not get as far as demolishing the boot files and the bike recovered 'undamaged'. This time the owner was not so lucky.

This all serves as a reminder that whenever tackling the 'unknown', first read (and follow) the ******* manual.

I hope that this script helps to explain what is going on / gone on.

"If in doubt, do nowt" (old Yorkshire proverb)

The Laird.
(Telling it like it is, as always)

I agree entirely with DaveW(having programmed a few pic chips etc in my time) that the firmware should be reprogrammable somehow regardless of how "bricked" it is.

Absolute worst case scenario - remove pic chips from board and reprogram

Chances of getting the information from Vectrix i reckon are about nil however, but if XVectrix would like to jump in at this point we'd love to hear from him.

Interesting (off topic a little) but I used to be into the Games console repair and mod business and one thing that was a major problem was people dropping Sony PSP's and somehow bricking the firmware (or doing a firmware update from sony and it going wrong). There was no known way of reflashing the firmware other than have the part shipped back to Sony for them to do it for several hundred dollars. This was until one day someone received their PSP back from sony with a different coloured battery in it. Long story short the battery contained a control chip that when one leg was shorted it allowed firmware from an SD card to reflash the PSP regardless of state of erasedness (is that a word). So systems are often there but unused and undocumented. Another well known one is what allowed Micr*s*ft to gain some degree of market share with their Xbox but thats another story entirely.

So.... come on vectrix.com or vectrix.com.au as you are not helping in so many other ways - give us this info and let your LOYAL customers fix our machines !

I'll certainly be interested in the process that they take - anychance of videoing it ?

This problem has been a PITA since day one, especially with the older diagnostics. The MC uses an older TI processor that does not retain the bootloader code during reprogramming. The architecture requires that all the flash be erased before loading new code. So, during MC reprogramming, the first thing the bootloader does is copy itself to volatile RAM on the processor. The bootloader then begins executing from RAM so that it can erase the entire flash, then load new code. Then it copys itself back to flash. The other issue is that the available RAM is so small that the bootloader code cannot be made very complex or robust. It literally uses every available address of RAM. If one line of code were added, one line would have to be deleted somewhere else. The result is that any glitch in the CAN communication or anything can cause the reprogramming sequence to fail or hang. Sometimes it is recoverable by restarting the diags, but it depends on where it was in the reprogramming process. Turning off the scoot leaves the MC in an unrecoverable state. Once this happens, the only way to recover is to reprogram the MC directly from the JTAG port on the PCB. Its very simple and fast but requires some expensive programming hardware. This puts the code and BL back in flash.
Most newer processors have flash that can be erased in segments. So you can put the bootloader in a protected area and not have to worry about losing it. Also, there is lots of code space available so it can be made very robust.

Once this happens, the only way to recover is to reprogram the MC directly from the JTAG port on the PCB. Its very simple and fast but requires some expensive programming hardware. This puts the code and BL back in flash.

Isn't JTAG a processor independent standard? If so then there are inexpensive JTAG programmers out there (the AVR Dragon is the one I use which costs ~$50). If JTAG is a standard and if it covers software as well as hardware then is it conceivable that any JTAG aware IDE could be used to reprogram the chip (not to recompile the code - just to pass a pre-compiled HEX file over the JTAG wires)?

I'm guessing the answer is no - but I figured I'd ask. If the answer is yes, and if the hex file is available (or can be downloaded from an existing bike?), then there's a $50 solution (the AVR IDE is free and includes a programming tool which can be used independently of the compiler/linker).

Also, what's the part number of the MC that needs reprogramming? Is it a popular MC? If it is then google is our friend...

Just my 2 cents worth.

Oops - won't be able to use the AVR IDE because it handshakes with specific Atmel processors.

Probably won't be able to use the AVR Dragon unless someone else has already figured out how to interface it to the micro in question.

It was worth a shot :-(

Also, what's the part number of the MC that needs reprogramming? Is it a popular MC? If it is then google is our friend...

Clarification - XVectrix used MC for the entire controller and I misread and used it to mean "Micro Controller". I'm interested in the part number of the "older TI chip".

p.s. David are we ever going to get an Edit button?

jdh..
You may be correct. I have not looked for a programmer in probably 6 yrs so I am not up to date on what is available and compatible with the TMS320F241 processor. Back then, the SIGNUM was the cheapest one I could find that was compatible. (~$600, back then)

Hello Folks,

I am having some guilty feelings

Laird, if it weren't for your software contributions, a lot MORE Vectrixes would be doorstops! If we had had access to your knowlege several years ago, I might still be getting 35 miles to the charge, instead of 15-20. At least with your new software, charging means cooling, so I can charge anytime, anywhere, without haveing to worry about delay times or cooking the battery. I LIKE IT! :-)

Hi Guys,

I'm not convinced the motorcontroller has failed here for several reasons - I'm relatively sure the ICM has failed instead.

A failed MC won't stop you from flashing the charger.....

A spare ICM may be easy to get hold of than a spare MC, I would try swapping a new ICM in before pulling the MC out

Getting the MC and ICM from New Zealand is probably your best bet, they aren't large so shipping shouldn't be too bad.

If I try to program the motor controller firmware, I get this message:
"No response to Clear .. Please try again"

And if I try to program the charger firmware, I get the this message:
"Did not receive DM15 for access address 0x200"

I can reprogram the ICM with no problem.

Looking back at the correspondence between myself and Scott, the no response to clear only occurred after the ICM was flashed. The charger had already been successfully flashed.

The MC flash was later attempted, but on the first try only got as far as sending the clear command with no response.

The flashing process for the MC won't actually start until the program has access to all 3 hex files: CLR24_X1.hex, ERA24_X1.hex and PRG24_X1.hex

The cure to Scott's problem is, more than likely, a Factory reinstall of the boot files or a replacement controller with the files already fitted and intact.

Between Hiiba and I, we have a fairly large pile of dead MC's, we could probably send over a micro-controller with the program still in it.

It would be easier than trying to get a new one, or even reflashing it.

But try the ICM first if that's easier to get hold of

I for one believe putting The Laird's firmware on the bike is worth the risk.

Matt

Quick google and....

http://www.keil.com/download/docs/7.asp

and then lo... there it was

on this page
http://e2e.ti.com/support/development_tools/code_composer_studio/f/81/t/68158.aspx
"there is a -b option built into the hex conversion utility from TI, so you don't even need hex2bin. However, you'll still need to have your code/data in one contiguous block. You can use the -image option to generate an image of your code."

and further down

"This TI hex conversion utility is called hex470.exe which can be used with the -b option to generate .bin file directly from .out file. It has additional options which may help you to configure your file as your requirements."

Good luck

I've started down the road of JTAG programming. I pulled the rear battery and removed the MC cover (only to discover that I could have accessed the JTAG port without doing any dis-assembly at all (there is an access hole in the MC cover .. Oh well, live and learn) The next steps are:

a) buy (or borrow) a USB->JTAG adapter (I found one on Ebay for $85)
b) Install the TMS320F241 development SW
c) the JTAG programming requires the data to be in a binary (.out) format rather than .hex format .. so I need to either get the .out format file from Vectrix or figure out how to translate .hex format into .out format

Yay, this is sounding like promising progress. One way of getting the binary format might be to hook up the JTAG device to a working bike - you should be able to dump the contents of the flash. Do you know anyone near by with a Vectrix?

Hi Scott,

did you make some progress with the programming during the last year ?

regards

Klaus

I am in the same situation with my MC.
After 70Ah conversion, I get the same message during programing new MC firmware.
I am looking for an second hand MC in good condition....

You and bm3 have bumped one of the weirder threads out there. SLbaker and jdh2550_1 just sort of vaporized mid-discussion like h-bombs were dropped on their respective towns. As a Current Motor scooter owner, I have a good suspicion why jdh2550_1 (CuMoCo co-founder John Harding) vanished. It was probably a gag-order as part of a less-than amicable departure settlement with the company. I wonder if SLbaker got gagged-for-money (or a working MC board) from Vectrix or The Laird?

Seems Scott was not successful in getting his machine running and is now selling it for $1,500.
http://portland.craigslist.org/wsc/mcy/4420544431.html

Hi I'm new to V is for Voltage, I hope this is an appropriate thread to post my plea ?

I've got a 2012 Vectrix VX-2 with less than 1000 miles on it.
After recent recharge the speedometer and energy gauge stopped working and instead only displays "Error System 01".
My dealer in Oakland, CA says he's heard of this before and sent me a partial quote from earlier e-mail from Vectrix before they closed their doors in most recent bankruptcy.
I quote: "An Error System 1 message has to do with the firmware coding in the speedometers, it does pop up sometimes and can be “stopped” temporarily with a specific combination with the throttle lever- I will send this to you shortly, but if this does persist, or pops up more often, you will need a replacement speedometer."
Does anyone know the exact procedure to reset or reboot the firmware ?

thanks in advance

Seems Scott was not successful in getting his machine running and is now selling it for $1,500.

How amazing. An entire machine in perfect order rendered an inert hunk by a software bug.

Has anyone attempted to simply convert a Vectrix to a "dumb" scooter using a different (aftermarket) motor-controller and BMS? The biggest challenge would be getting the gauges to work.

Was a service manual with wiring diagrams ever made available for the Vectrix?

VX-2 with less than 1000 miles on it.
After recent recharge the speedometer and energy gauge stopped working and instead only displays "Error System 01".
My dealer in Oakland, CA says he's heard of this before and sent me a partial quote from earlier e-mail from Vectrix before they closed their doors in most recent bankruptcy.

A Vx-2 is a rebadged E-max, a very different bike to the other vectrix bikes

Emax are likely still around, being part of the larger V-moto group - they should be able to point you in the right direction

The VX2 and the Emax share the same frame, instrument cluster (with new faceplate) and motor. Altho the frame was modified somewhat for the Vectrix body work and Vectrix drive sysytem. The original code for the dash was developed for EMAX by an Italian company. The code was then taken over by Vmoto in China when Vmoto started making the VX2. Vmoto only had one guy who knew anything about the code. Vectrix was able to get a few changes to the code before he left Vmoto.

The sequence to clear the error is this:
Key off
Kill switch off
press right brake
key on and hold 15 seconds till error clears.

Hope this helps

Thanks so much for the background info, and for the procedure.
I will try it today and let you know if it works for my scooter.