Wednesday 9 December 2020

CNCdrive USB programming stick - shagged?

As noted on Sunday, I seem unable to talk to the CNCdrive DG4S16035 servo drives on The Shiz. I have the USB programming jobbie but although it's recognised as a USB device by my laptop, there's no visibility of the drive beyond that. Seems the USB controller can't talk to either of the drives I tried.

I tried to order another one from their web store but as I've seen before, you can't actually check out and pay for stuff on the website itself. Instead, you have to wait for them to get back to you for payment details. 

That's Noddy enough as it is but (as I've also seen before) they don't seem to be able to bring themselves to even respond if you are only after a piddling little order. It may only be a small part but if it stops your whole system, that's not a very mature attitude. But what would I know - perhaps that's how real men behave in Hungary.

Bollocks - there's not much to go wrong on this thing. Assuming I haven't done anything dumb worse than plugging it in backwards, it's not easy to see how it can be anything other than a popped component. 

Seems to me I could have plugged the fucker in with the power on or possibly zapped it with ESD. Or plugged it in backwards etc.

Let's have a look at this thing:


The iPhone struggles with something this small but there are only 2 devices there plus a few Rs and Cs.


The 8 pin device is a Texas Instruments ISO7221A digital isolator that simply provides isolation between the PC and the drive. Then there's an FTDI Chip FT232R USB UART IC that does the USB thing.

Not obvious how I could have shagged the USB IC, given that the isolator should protect against misconnection of the rather home made 4 pin header thing. And indeed, the USB IC must surely be functional, as the laptop was "seeing" a USB device.

Most likely is that I've managed to bugger the digital isolator, most likely by plugging in while powered or perhaps misaligning the pins. 

The absolute max voltages for the isolator say -0.5V on any (either) of the inputs, so reversing or hot plugging them could easily knacker them.


Sod it, it would be quicker and easier just to buy some replacements from RS than wait for these guys to get off their arses. CPC don't stock these devices and Farnell wants £10 "handling fee" - plus P&P. RS don't do that nonsense and the (next day) P&P is free for >£30 orders. RS it is.

By the time I've bought the MOQ of devices and some right angled pin headers, I'm a few quid short of the £30, so why not get some constant current diodes LM334 so I can drive LEDs from a range of voltages - useful for debugging the LinuxCNC controller perhaps. That takes the total to about £31. Job done. Now get on with something else until this lot arrives tomorrow.

No comments:

Post a Comment

Final assembly and test of the spindle nose adaptor - RESULT!!

After the recent distraction caused by the 3D scanner, resurrecting the 3D printer and buggering about with the throttle bodies for my Honda...