Mobility Scooter Power Supply Repair.

Thread Starter

studiot

Joined Nov 9, 2007
4,998
Any experience fixing an Pihsiang MB-24/3 mobility power supply would be gratefully received as new ones are exorbitantly priced.

These units are for charging the 24V batteries in mobility scooters.
The faulty one appears to have an intact transformer with a CT 60-0-60 main secondary and a much lighter 10-0-10 for the control electronics. The main 20A output fuse is also intact.

The control appears to be using two BT151 thyristors in a biphase rectifier (not smoothed) arrangement with the main secondary.

The control circuitry comprises two HA17324 ICs, which I think are equivalent to LM324 quad op amps.There is 18 pin DIL IC which I do not recognise, labelled CF74S an ??47P. I think this is a PIC processor.
 

R!f@@

Joined Apr 2, 2009
9,918
labelled CF74S an ??47P
Get a clear pic of the label. May be I can make it out.

The thyristor is labeled BT151?
Dismantle it and give me a detail of the gate drive IC
 
Last edited:

Thread Starter

studiot

Joined Nov 9, 2007
4,998
Yes the scrs are labelled SCR1 and SCR2 on the board and BT 151 on themselves.
Here is aother pic of the board with the cables drawnout of the way plus a closeup of the driver.
 

Attachments

R!f@@

Joined Apr 2, 2009
9,918
Anyways the CF Chip might be a PIC.
the other two's are 324's.
can u get a pic of the track layout and check the fuse + SCR for leaks between gate and anode.
Do a quick diode check on the transistors.
 

Thread Starter

studiot

Joined Nov 9, 2007
4,998
The main 20A output fuse is also intact.
I haven't done any checking except to see that the DC supplies are what I expect.

There are 4 ICs

The TO92 item U1 by the driver chip is a 78L05
U2 & U3 are the op amps
U4 is the driver

There are three transistors and sundry diodes.

I will have to physically remove the board, which means some unsoldering, to get a connection pattern.

I stopped before at this juncture as I see no point unless I can prove the driver chip is working, as I cannot replace it if it is a special.

I do however have a second working supply to compare readings from.
 

Thread Starter

studiot

Joined Nov 9, 2007
4,998
I didn't mean I was giving up, just that it will take a bit of time to get the board out.

Also that I really need to identify the type and function of the driver.
 

eblc1388

Joined Nov 28, 2008
1,542
The CF745 is actually a version of 16C54 which is only available in Asia. The link of the actual CF745 PDF always redirects me to the actual 16C5X datasheet so it could be the same as 16C54 albeit by another name.

This is an image I got from a Chinese Vendor. The blue explanation is added by me.

Hope this helps.

 

Attachments

Thread Starter

studiot

Joined Nov 9, 2007
4,998
Thanks eblc, that could be really useful information.

I had been hoping that some member might have fixed one of these before and could provide some clues.

Never mind, I will keep posting as I work through the repair. Perhaps the information will be of use to someone else in the future.

Meanwhile I have removed the front panel and drilled out the concealed rivets, releasing the board so here is the trackside. The BT151 SCRs are at the bottom of the pic.

The makers really didn't want others to dismantle this box.
 

Attachments

eblc1388

Joined Nov 28, 2008
1,542
Studiot,

Based on the images you have posted, I have made you an aid which will greatly reduce your time in tracing up the circuit. See attached.

Note the PCB traces shown is what you would see if you have Xray eyes from the component layer.
 

Attachments

THE_RB

Joined Feb 11, 2008
5,438
Please explain the nature of the fault, ie; what it did when it failed, what it used to do when working and what it does now when broken.
 

Thread Starter

studiot

Joined Nov 9, 2007
4,998
Please explain the nature of the fault, ie; what it did when it failed, what it used to do when working and what it does now when broken.
The first photo post #4 says it all.

There is a cable with an XLR plug to connect the charger unit to the mobility scooter.
Normally the lights work as indicated in the picture and the battery charges.

With this unit the red fault light flashes and the battery does not charge.
 

Thread Starter

studiot

Joined Nov 9, 2007
4,998
So far I have removed and tested the three transistors, two SCRs and one 7805L regulator and the transformer. They all appear functional.
 

loosewire

Joined Apr 25, 2008
1,686
you did not say you changed the back up supply.The board shows the plug ins.You said you desolder several parts,sound like too many with out good reason.You did not show the voltage readings on the good supply. Did you load test the battery,maybe you have a bad batter that won't charge,I'm surprised that some of the old regulars are not in this conversation.You hear all the time that a lot of solid state parts
don't go out ,with out the smell,touch,sight test. I'm Loosewire, you need
more input before you have two supplys out.
 
Last edited:

eblc1388

Joined Nov 28, 2008
1,542
So far I have removed and tested the three transistors, two SCRs and one 7805L regulator and the transformer. They all appear functional.
Can you post the info on what part number is the three transistors Q1~3 and/or they are NPN or PNP?

You've said the red LED is flashing. This is not bad news after all.

The LED is controlled and flashed by the controller CF745. This means that the PIC is 99.9% functional as it must be in order for the LED to flash, unless the designer had chosen to use a self flashing LED which I doubted.

The designer has used the PIC to select charging modes, charging voltage/current limits and monitor the function of analogue parts. These are low voltage/current signals and rarely has the chance to damage the PIC.

The actual firing of the SCRs are provided by the output of the HA17324 and there is a zero crossing functional block formed by two transistors at the top left part of the PCB next to the three rectifiers.

There is a cable with an XLR plug to connect the charger unit to the mobility scooter.
I hope you have already eliminated the possibility of a loose or broken wire regarding the plug and have checked that all its connections are good.

Edited: added transistor part/type request
 
Last edited:

Thread Starter

studiot

Joined Nov 9, 2007
4,998
Thanks for your continued support, eblc.

Yes I have checked the leads for connection and continuity. As I also mentioned I have another working unit so I can now start comparing voltages and signals. This second unit charges the scooter fine.

The three transistors are all the same general purpose NPN type 2SC1815.

Yes I'm hoping that the PIC is not damaged.

The next thing (after putting the other semiconductors back) will be to look around the op amps to try to test them without removal.

One other note, there is no sign of damage, burnout etc around the other components, including the high wattage ones.
 
Top