E V Forum
Register
 
 
 


Reply
  Author   Comment  
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #1 
Hi to all who read this,, re the above,, .I have successfully made up a few of the usb/serial interface cables and all work well with the older curtis software, both with windows xp and windows 7,,, If any one needs one, message me on the forum and I will send you one at cost,,
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #2 
Hi,
Does your design output +/- 12 volts to the controller?  I have a similar USB - RS232 adapter, but it only produces +5/0 volts, which I think is insufficient for the Curtis 1228 controller.

Thanks in advance.
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #3 
Hi, as far as i know the output is +5v,, it uses an arduino board , I have a 1228 being delivered sometime this week and will try out my interface on it and get back to you,, Regards,Ralph
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #4 
Hi, I had the 1228 delivered today and have just tested my interface,, Happy to say that it works perfectly and I am able to program using it,Regards,Ralph
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #5 
Hi Ralph,
Hmmm, interesting.  That means that the controller I'm trying to fix has at least one more fault!  The Curtis program fails to make a connection, with a variety of error messages.  Scoping the signal at the Rx/Tx pins on the PIC shows only 3 volts.  My expectation that 12 volts is needed comes from a few internet comments, plus the Curtis programmer takes the 15 volts provided at the 4-pin molex; I wonder why?
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #6 
Hi, What I am about to say you will have to make your own mind up about,, bearing in mind that I am no electronics wizard, only a guy who researched the method of constructing the interface,, Firstly, If i read your post correctly you appear to be using all four terminals on the 4 pin molex, when in fact ONLY 3 are used,, TX,,RX,, and ground ,the 12 volt terminal is redundant,,
Also, Are you remembering that the controller needs to be powered up ?? ( +24v on pin 5),, .
Good Luck with your repair
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #7 
Hi,
Thanks for taking the trouble to respond.  I have quite a bit of experience in electronics; when I started TVs were B&W, phones were big, black things - assuming you could afford one, and the only electronics in a car was the radio, again if you could afford one.
You must know something - I'm intrigued by the use of an Arduino as the 'interface'; never had a play with one of those, but I did have an encounter with a Raspberry pi.

With the Curtis controller I suspect the PIC has died, although it shows some signs of life. One of my problems is that I'm not sure how a working unit would behave; applying 24V nothing happens at all, but attaching the diagnostic LED shows it to give some flash codes, but none that make any sense!
BTW If you have a faulty 1228 controller I would be interested in buying it, so I could do some 'comparative anatomy'....

Thanks again.
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #8 
Hi, the flash codes can easily be found on the curtis website,, just go there and virtually all the controller info (apart from schematics sadly) can be downloaded, however if you are getting flash codes it would suggest that there isnt too much wrong with the controller,, I have found that the usual suspects for the fault codes are missing/open circuit brake solenoid, or wrong type of control pot One thing to be aware of is that the controller dosnt stack the faults,, so one has to be cleared before a possible second can be seen,, If you take a look at the "buggies gone wild site, you will find the info on making up the interface,, it as i mentioned earlier uses an arduino uno board and a chip which is connected (which i believe inverts the transmit and receive signal,, If you cant find the info let me know and i will dig out the info,,
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #9 
Hi,
Thanks again for the help.  I've just about squeezed all the useful info from the Curtis site (wouldn't a schematic be nice? Perhaps someone with a lot of time to waste could reverse engineer one)

When connected up to the buggy and switched on the controller only lights up thp  BDI, showing full charge, but does nothing else.  So I've rigged up a speed pot, dummy brake, along with the LED.  What's puzzling me is that the flash codes are not as given in the manual; usually there are three different sets of the 'short' type, with the last one continuously repeated for 'throttle not at zero' - which it is!
When you say the fault codes need to be cleared, I assume this is done by cycling the keyswitch?
To add to my confusion, occasionally when turning on the keyswitch the unit will show some signs of life - the relay closes, the LED stays ON, but varying the speed pot. shows no activity on the mosfet gates or drivers for the pwm drive.
Hence my conclusion that something somewhere is fritzed!
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #10 
Hi, Just a couple of pointers,, certain  faults can be cleared by recycling the power switch and must be rectified or they will return they  remain in memory till cleared with the programmer,, and as I mentioned earlier, faults arent stacked so should  only show first fault code.
A word of warning,, make sure the "dummy brake " has the correct resistance or if not it may blow the mosfet, (this from previous experience sadly).
Are you also aware that the accelerator pot needs to be correctly programmed for type ?? once again this is done via software and could be why nothing is happening,, also the pot must be between 4.7 and 5.7 Kohm or it faults the controller.
What flash codes are you getting at power on ??, and why do you think they are wrong ??,, hope this will help,, Are you still not able to program ?? cheers
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #11 
Hi,
I think I have done all the right things in my test harness, although I have assumed the speed controller is the simple 5k pot. type.  The flash codes puzzle me since on turning on the keyswitch it produces 4 different sequences, some of these are given in the manual as 'slow' codes, whereas in fact they are all 'fast'. The first three are meaningless, i.e. not in the book, but the last one repeats continuously as 3 - 1 for pot. not at zero. All of this could be made more clear if it were possible to program to controller; so far all I get are various error messages, all of which simply mean that no communication with the controller is possible.  E.g. it gives 'checksum error' , 'comm. error' , 'timeout error' which is odd, because scoping the Tx line shows that no response is ever made by the PIC!
All of this leads me to the conclusion that the PIC has been fried; I have some experience of Microchip PICs and have found that faulty ones can seem to work OK, but produce all manner of weird behaviour.  I asked Curtis for a replacement PIC and got a polite 'get stuffed'!  I have a couple of 16C74's so all I need now is the firmware to program one!! (FYI the program data must be stored in the I2C EEPROM, as the PIC is a one-time-programmable.)
Thanks again for your patience...
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #12 
Hi, I doubt that I can be of any further help with this,, There is just one possibility in that since you are getting the "pot not at zero" fault, it may be that the deadband and throttle gain settings have been set up with a 5k pot with a greater range of travel .(most proprietry throttles, operate on around a 90/100 degree rotation from zero to full output) also, have you remembered to jumper pot high to the speed limit pot output (assuming you arent using a speed limit pot),,, .
You didnt say if you have reproduced the interface ,, (using thee arduino) ??, may be worth a try,, cheers
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #13 
Hi Ralph,
I'm at a bit of a dead-end now with this one.  Although I feel sure that I'm doing the right thing, always at the back of my mind is that I'm missing something important!

Anyway, your original posting was about your USB/Serial interface, and I would like to get hold of one, as you say it works with the 1228 type controller. Am I correct in assuming it operates with the Curtis 1314-4401 PC program?  Presumably I need to send you a PM?

Regards.
0
Ralph Bolton

Junior Member
Registered:
Posts: 8
Reply with quote  #14 
Hi, firstly in answer to your question, Yes, it runs on the 1314 v3.12.
Having given it some thought I have decided to make you an offer,, The 1228-2701 which I bought last week purely as a standby for my golf buggy, I am prepared to sell for what i paid for it,and would also send you a working interface , that way you can do the comparison between yours and a fully functioning unit,, and of course perhaps solve the puzzle of your own unit.
The unit cost me £90.00 bought from a local mobility dealer and the cost of putting together the interface was £16.00 ,, If it is something which would interest you I could put both items on ebay ,, the postage would be around £7.00 and so that I dont loose money I would need to add a further £5.00 for paypal and ebay fees,, bringing the total to £122.00,, let me know what you think,, regards 
0
mikew8760

Junior Member
Registered:
Posts: 18
Reply with quote  #15 
Hi all,
Thanks for all your help, Ralph.
After a fairly long process, with time out for holidays, I have finally come to the conclusion that the 16C74A PIC in my controller is faulty, and the rest of the board is OK.  I got a cheap-ish controller, similar to the 1228, but had to hack all the rubber sealant away and fit a Molex connector, but at least it got the buggy going again!  (To my disappointment, the PIC was soldered in on this boards, whereas mine is socketed!)
Curtis won't sell me just a PIC - no surprise there - "we don't supply spare parts".  Obviously the PIC will have been programmed with all the code protection fuses set, so no possibility of cloning one.  But I wonder, since there is the serial comms available for modifying the controller parameters - which I assume are stored in the serial EEPROM - perhaps Curtis also included a 'back-door' so that, in fact, the code may be retrieved? 
Has any expert guru on Curtis products any knowledge if such exists?

Regards.
0
Previous Topic | Next Topic
Print
Reply


Create your own forum with Website Toolbox!