WheelchairDriver

Made up a programming cable per this illustration with a USB-serial module. The one I had on hand is CH340 based but the operation is the same as the FTDI FT232L based units. I use the CH340 units in several applications, mainly amateur radio digital modes, have found them to be very reliable.



Connected the TTL output gnd to pin 1 on the 4-pin Molex, data to pins 3 & 4.



No problem getting my USB-TTL module recognised as a COM port. The port settings are 19200, 8N1.



When I run the mfr programming utility, we can see activity on the data line. Sending a packet like this repeatedly, probably trying to wake up the S-Drive unit.

When I run the mfr programming utility, we can see activity on the data line. Sends a packet like this repeatedly, probably trying to wake up the S-Drive unit.



The scooter is powered up, but the controller does not want to talk to the programming utility. COMMS INACTIVE message.

Have tried the cable with a 47K, 33K & no resistor.

Any hints? Should the port be set for a different speed?

Can't get 'em to talk.

Thanksabunch

-Brian

Have YOU told the software what Port number your CH340 is using?

NOTE:- This assigned number MUST be between 1-8 inclusive.

EDIT ...Your CH340 appears to be assigned COM PORT 8 ....have you told the programming software that it should look at COM Port 8?

Tools ...Comms ...select.

Yes, the software is using COM 8. It's the only serial port on the machine. Otherwise, it would throw a 'port not found' error & not send the data that I'm seeing with the sillyscope.

Try with separate TX and RX lines... NO resistor or diode required....... as I recall pin 3 of the Molex is RX and pin 4 is TX.

Ah! That sounds like a better way to go. That's what you typically see in these sorts of applications. Will have a try in a few.

Thanks!

-Brian

Unfortunately, I could not get the S-Drive to talk to the PC.

At the same time, I appear to have induced a problem. Now the scooter is stuck in charging mode with no charger connected. The LED bar display constantly scrolls and the scooter will not move.

I have gone through the tiller circuit board, replaced a couple of 2SD313 transistors (probably pass transistors for the turn blinkers) because I had them on hand, checked a couple of 2SD4091, both good, and replaced the LM324N quad op amp (because I had them I could not test the one on the board in circuit), but still have the problem. All I can figure is that I somehow hurt the speed controller.

Any advice?

Thanks

-Brian

Seems that you have caused an inhibit condition inside the S-drive ....possibly one of these ( not sure which type they are ) has blown and shorted to ground.

You will find them / it on the S-drives board connected as I recall to the Molex RX - TX.

Excellent, Woody, thanks for that. Most appreciated.

Cheers

-Brian

It may be possible to reverse the inhibit, or disable it in programming. It is with most chairs. But never looked at any scooters. But... Err... You cant!

Thanks for that!

There's a paucity of data around for the S-Drive controllers, no doubt by design to protect IP. A pinout would be nice, a schematic & parts list for one would be even better.

Cheers

-Brian

OK, easily found, catastrophically cooked a SOT-23 xistor.

Now to find out what it is (was) because there's not much of it left. Very hard to photograph but may have been a silver topped package with a type number beginning with 'B'.

As if that narrows it down a lot.

I'll open one up and have a look ...

VND5N
Image from web...

DO NOT PLAY WITH OR USE THE POSITIVE 24V at the Molex.


https://uk.farnell.com/stmicroelectroni ... dp/2629755
https://www.digikey.co.uk/product-detai ... ND/2529391

OK, that's not the part that's cooked. It's the SOT-23 device northwest of it.



Thanks much for your help!

OK...I will need to drag out the microscope for that.

As I recall it's essentially 2 diodes end to end.

BRB

BA25 ...Pics of before cleaning and after.

Multimeter suggests that it is just a diode.

So ...you could try this ..

Excellent, thanks for that. Over & above the call of duty, Woody!

Searching for BA25, I find a ROHM low drop out 2.5V voltage regulator by that designator in a variety of packages, current sourcing increasing with package size, but none in SOT-23.

https://au.rs-online.com/web/c/semicond ... hTerm=BA25

Really weird, not used to not being able to easily find a component by the printed designator.

Getting a bit late here in eastern Aus, but in the morning I'll try to sus out more. If it's a vreg, there's a few ways to skin this cat.

The scooter is wired such that if a charger is plugged in to the charge port, movement is inhibited, whether the charger is on or off. I believe that's because the inhibit line is tied to gnd in the charger. If that's the case and the component (or remains thereof) on my controller is shorted, then removing it may resolve the inhibit condition. If I do need a 2.5V signal, I can certainly find another vreg to do that even if I can't find the part the OEM used.

Thanks so much for your help, Woody!

Cheers

-Brian

Doubt it is reg ...it's only connected on 2 legs and multimeter suggests it's a simple diode...unless the other leg is connected thru the board.

http://www.mouser.com/ds/2/348/dan217-210030.pdf

Ooh, good score. Sometimes a Google search can lead you down the rosy garden path. I'll poke a 1N4148 in there & see how we go.

Really appreciate your help, Woody!

Cheers

-Brian

I removed the remains of the burned component. Found that there's 3 lands for the device, so whatever it is, it's a 3 leaded device, whether a transistor, vreg or perhaps a dual diode.



While I had no component in the position, I connected it to the scooter. The bar display began blinking. 8 of the LEDs blinked 8 times in a row. Presume this is an error code, but I haven't got a list of them.



I put in a 1N4148 per Woody's suggestion.



This returned the behaviour to a scrolling LED pattern which is normally displayed when charging.

Hoping that BA25 might be a dual diode package and that one more diode might be the fix.

Thanks a bunch, fellas!

Cheers

-Brian

Oh, hang on, it is a dual diode package (duh, Brian, look at the datasheet )



I'll go rework it with 4148s per that pinout & see if that sorts it.

Put two 4148s on the lands per the line card (not really a datasheet) illustration.
Nope. 8 Bar flash code.



Sensing that since I'm looking at a line card & not a specific data sheet (still haven't found a datasheet for BA25 in a SOT-23), I swapped the orientation of both diodes. Same, 8 Bar code.

Welp ¯\_(ツ)_/¯

I'll probably keep trying to repair this controller to have as a spare, but I really need this scooter to work, soon. I am stuck in the house without it & am dumping all the shopping, errands etc on my beloved Trophy Housemate.

Wheelie Good Mobility in Kent UK had several of these controllers on hand, so I've ordered one in.

#sigh

Thanks for your help anyway, Woody, most appreciated.

Cheers

-Brian

The diode orientation that I posted was correct according to my multimeter test of an intact BA 25 ...thus the other is as per the pdf diagram..you should put it across the Molex pins.

This is ...I.M.O Correct ...

Thanks much for the schematics. I've been diligently squirrelling those away as I've encountered them on this board but the more places they can be found, the better.

Unfortunately, there appears to be more bad stuff happening on this controller than a blown dual diode package.

#sigh

Thanks again!