DIY Electric Car Forums banner

1 - 13 of 13 Posts

·
Registered
Joined
·
5 Posts
Discussion Starter #1
Hi, I'm a noobie here, so please be patient!
I have an old golf buggy with a Curtis 1228-2730 which I use as a mobility scooter around the paddock. It has always been a bit 'intermittent' sometimes it just played dead.
After much faffing about I have discovered that the EEPROM is faulty (not too difficult as my background was in electronics).
The problems now are, a) if I replace the EEPROM (93LC86 for those playing along at home) with a new one, will it be possible to re-program it via the Curtis PC 1314-4401 software, or b) acquire the correct content and program it before fitting? (From where I've not a clue!).

I'm hopeful that a) may be possible, since there are no ICSP or JTAG connectors on the board, and I think it would be unlikely the EEPROM was pre-programmed in manufacture.
Perhaps someone more conversant with these controllers can put me wise?
 

·
Registered
Joined
·
5 Posts
Discussion Starter #2
Hi, it's me again...
Well, I've replaced the faulty eeprom, programming it with what seemed to be rubbish taken from the faulty one. Of course, now I get the 4/3 LED flashes, indicating an eeprom fault. The manual says how to clear this fault, but now the controller won't connect with the programmer. There's quite a lot of back-and-forth data comms (at least, the led's on my USB to serial converter flicker a lot) but after about a minute of this an error message pops up.
I'm suspicious this is now due to incorrect data in the eeprom, but with no data comms it's impossible to correct it; Catch-22?
Any bright ideas, folks?
 

·
Registered
Joined
·
8 Posts
Did you ever sort it out? I have pretty much the exact same controller with exact same issues. Intermittently reports EPROM failure. Downloaded the manual and firmware update + program, but no joy. Not even sure if it is accepting the programming. Using the RS232 port btw
 

·
Registered
Joined
·
5 Posts
Discussion Starter #4
Hi,
The short answer is No! The slightly longer answer is that after much time spent attempting to diagnose the problem I came to the conclusion that the PIC, 16C74 microcontroller was faulty. I may be wrong here, but I noticed that there had been some arcing on one of the 24V spade connectors on the controller; voltage spikes are a good way of killing such devices. There is, of course, no chance of obtaining a spare or the firmware to program a new one. (Unless you know better!?)



You say you are not sure if the RS232 serial link is working. If it isn't the PC program will soon tell you!! The laptop I use only has USB, so I use a USB to Serial converter, but the operation of the program (Curtis 1314-4401) would be the same.
Sorry I can't really help you further; I've spent many hours and set up a test rig to probe the damn thing, but failed to nail the precise nature of the fault.
 

·
Registered
Joined
·
8 Posts
Thanks for your comments Mike. My story so far -
A friends fathers mobility scooter stopped, which I found to be moisture on the pcb.Stripped it down, cleaned and checked all pins / signal levels. All looked good. Powering up / down several times gave different results?. The main one was error code 4F+3F = EEPROM Fault. Sounded legit.
Made up the test rig (attached), found the Curtis "PC Programming station" software, with latest firmware. 1314-4402 ver 4.6.7 and tried to re-flash it. Nothing. Tried other serial ports ... nothing.
Checked these forums and it seems others who connected RS232 signals direct, blew up their serial ports? Checked 2 of mine, yep both HOSED. Traced the circuit, and both lines go to a DS2003, which has +-50V tolerance, so do RS232 ports, so it doesn't make sense. There are no shorts to any other part of the circuit.
A search here suggest that you convert to TTL levels and INVERT the signals !!?? I'll try that next I guess. Haven't found anyone with a success story . POS is not going to win :)

I'll re-post with progress.
 

Attachments

·
Registered
Joined
·
1,478 Posts
Were you able to read the configuration register byte to determine if code protection was ON or OFF? It is off by default on a blank chip and must be programmed at 2007h. If it wasn't then we could reverse the firmware code.
 

·
Registered
Joined
·
8 Posts
Hi,
You say you are not sure if the RS232 serial link is working. If it isn't the PC program will soon tell you!! The laptop I use only has USB, so I use a USB to Serial converter, but the operation of the program (Curtis 1314-4401) would be the same.
Sorry I can't really help you further; I've spent many hours and set up a test rig to probe the damn thing, but failed to nail the precise nature of the fault.
I had written a detailed reply, and follow up posts / IMs, but not sure why they were all deleted !? I'll try to re-create as much as I remember -
Fault: My unit gives randomly - EEPROM fault error code 4+3, or a poor attempt to run the test motor. No rhyme or reason.
I have made up this test rig, downloaded and installed their "1314-4402 PC Curtis Programming station -OEM v4.6.7" software, d/loaded latest firmware and tried several times to "connect". All failed.
Doing more searching, I came across members who discovered that their serial port was damaged .. checked both mine and yep, both blown !! odd, since both pins go to a ULN2003, which is ~+/-50V tolerant. Luckily, it was only the RS232 i/face ICs, which I replaced. More searching, and it seems the trick is to use a RS232 > TTL converter PLUS Invert the logic?. I'll try that next.
Ken: In my case, I haven't been able to get ANY response from the controller yet. TBC
 

·
Registered
Joined
·
5 Posts
Discussion Starter #8
Hi,
It would be just too good that Curtis omitted to lock the code so everybody could extract it: of course, the contents are unreadable!

It has even crossed my mind to create a program that would allow the unit to control the motor, leaving out all the bells and whistles that Curtis include for various reasons. I even got as far as buying a 16F74 PIC, but that was all...



If you are that interested, my experiences with the unit can be summarised as follows:
The unit had shown intermittent operation, and then seemed to 'die' completely. But attaching it to the Curtis 1314-4401 PC software it behaved normally. But, when I attempted to change one of the parameters it went all doolally, and nothing would work. The LED flash codes 'suggested' an EEPROM fault, which made me think this was the root cause of the problem. However, changing it for a blank one only made things worse, if that was possible. The documentation for the Curtis programmer suggests that if a parameter is found to be out of spec then it will correct it. Since the 16C74 has no EEPROM, all the parameters must be in the 24C86 EEPROM, but which then creates a potential 'chicken and egg' problem - how to re-program the EEPROM contents in a unit that is unreponsive? It is most unlikely that one is pre-programmed prior to being soldered onto the pcb, which leaves the question, how is it done? I'd guess there is some part of the PIC that may be accessed by some secret code that allows the EEPROM to be filled with all the relevant parameters. Without this being done the unit just plays dead.
As you see, I've spent too much time trying to resurrect the damn thing, but without any real success. If you have any ideas please let me know.


Regards, and a Happy New Year.
 

·
Registered
Joined
·
1,478 Posts
Seems like you have the top-level programmer device which should be able to program a controller for any motor. Should be able to either read the parameters already in the controller, or set new parameters in a blank controller.

Alden DANG of nocoev com has posted some videos of reading controllers returned by customers--skip thru half as he unpacks the carton. Need to find someone like that who understands the curtis system.
 

·
Registered
Joined
·
5 Posts
Discussion Starter #11
Not clear which 'programmer' you are referring to. I only have the Curtis 1314-4401 after sales programmer, which uses the 4-pin Molex connector with a RS232 protocol interface. I use a laptop with a USB to RS232 converter (plus a data level inverter, as is required by the Curtis software.)


I know it is possible to access the EEPROM contents using Hyperterm and some encoding tokens I found on the internet, but it seems that without placing ALL the correct data at ALL of the correct addresses the system still refuses to play ball. (If you're interested in chasing down this particular rabbit-hole I can reveal more!)
 

·
Registered
Joined
·
1,478 Posts
i have an interest in how to re-configure a Curtis AC drive, so have been researching the options for using either the proprietary tools and software, or some bootleg-hockey reverse engineering.

As i understand the 1314-Xxxx programmer comes in 4 flavors that allow different privilege levels for access. the -4xxx is the highest (OEM) and the -1xxx is the lowest user level. That tells me that curtis is more interested in making money and control than sharing the operating system.

We were able to reverse the TCCH chargers on this forum, so feel free to share as much information as you have in as much detail as possible.
 

·
Registered
Joined
·
1 Posts
Hi, I'm a noobie here, so please be patient!
I have an old golf buggy with a Curtis 1228-2730 which I use as a mobility scooter around the paddock. It has always been a bit 'intermittent' sometimes it just played dead.
After much faffing about I have discovered that the EEPROM is faulty (not too difficult as my background was in electronics).
The problems now are, a) if I replace the EEPROM (93LC86 for those playing along at home) with a new one, will it be possible to re-program it via the Curtis PC 1314-4401 software, or b) acquire the correct content and program it before fitting? (From where I've not a clue!).

I'm hopeful that a) may be possible, since there are no ICSP or JTAG connectors on the board, and I think it would be unlikely the EEPROM was pre-programmed in manufacture.
Perhaps someone more conversant with these controllers can put me wise?
Hello, I have read the 93LC98 eprom of a curtis 1228, I can pass you the dump in hex format, in case someone may be interested.
 
1 - 13 of 13 Posts
Top