Slim Devices Squeezebox V2 with blank display

[anachrocomputer]

I'm helping a friend with a Slim Devices Squeezebox V2 -- it's a networked music player. It has a very nice 320x32 pixel dot-matrix vacuum fluorescent display (VFD) on the front, and no other activity indicators. The VFD is blank, though. We found two test points on the back of the VFD board, one marked EF and the other ANODE (J16). Assuming that EF is the filament, we found about 4.5V there, which seems OK. But the ANODE test-point we expected to be nearly 100V. We found 1.2V.

Does anyone know the correct voltages for those two test points on the VFD PCB?

Even better, is there a schematic available for the VFD board? We can see a row of capacitors and diodes that looks like a voltage multiplier, and a ferrite cored coil and a power transistor nearby. We think that's to step the 5V power supply up to the 100V or so that the VFD needs.

[pyoorkate]

Anachrocomputer mentioned that he forgot to take photos of the board whilst we had it in bits; given that it's the work of a minute I thought I'd quickly do that:



Just in case they're helpful. Sorry it's not quite as nice / cleanly lit as when he does them! 

(There's a few more in my photostream here: https://www.flickr.com/photos/49965961@N00/sets/72157629906324984)

[tom66]

If it's a standard VFD module bought in (possibly semi-custom) there's usually a RESET or POWERDOWN line that is required to be active before the anode supply will turn on.

[anachrocomputer]

We repaired the Squeezebox today! In case anyone else has a similar problem, I'll describe how we did it.

The VFD on the Squeezebox version 2 is a standard Noritake module that needs a 55V anode power supply. This anode supply was at zero, so we knew roughly where to look. It's a switch-mode supply that takes 5V and uses an NJM2377 SMPSU chip and a BSP296 MOSFET to generate 55V. We put the scope (Tek 5440) on the drain pin of the FET, which was at 5V with no switching. The gate pin also showed no switching. The oscillator section of the NJM2377 (pin 7) was working at 250kHz, and the voltage reference was OK at 1.5V (pin 8 ). We traced the circuit on the PCB and noted some resistor values. We began to look at sources for a  replacement chip and MOSFET.

Something odd happened, though. The bench power supply (Farnell D30 2T) made a brief buzzing sound when the device was switched on, for maybe a quarter of a second. This made me think it was getting overloaded (it's 2A and the original Squeezebox PSU was 1A). So we held the scope probe on the MOSFET gate again, while switching on. We saw a brief burst of switching then nothing. Aha! the NJM2377 is sensing an overload and shutting down! Now we checked more carefully around the output side, where the VFD's anode is connected.

We found about 120 Ohms between the anode pin and ground, and the resistance changed slightly when we reversed the multimeter leads (HP 3466A). I suspected the VFD at this point, but Kate suggested that we desolder a tiny surface-mount ceramic capacitor (C2) that's across anode and ground. Sure enough, the ceramic capacitor was partially shorted. We put it on the capacitance bridge (HP 4260A) and it balanced at about 33nF. The DMM (Vichy VC99) could not read it at all. So, having found a 40nF disc ceramic in the junk-box, we fitted a temporary replacement. The Squeezebox powered up with a splendid display of VFD graphics!

So, the lessons here are: always check the simple stuff (excessive load on SMPSU) and listen for even the tiniest clues (bench PSU overload buzz).

[eas]

Thanks for posting this. I picked up an old Squeezebox for $10 a little while ago, but hadn't gotten further than scraping the damaged polyurethane coating off the case plastic, and opening it up to look inside.

[anachrocomputer]

Glad you liked it! I've also written up the repair, with some photos, on my blog: http://anachrocomputer.blogspot.com/2015/01/repairing-squeezebox-v2-media-player.html

I've never seen a ceramic capacitor become resistive before; has anyone else here seen that?

[tom66]

I've seen one become leaky when hot. Cooling it down  made the circuit work.

[edpalmer42]

A couple of decades ago I had an RCA TV that wouldn't stay off.  I'd turn it off with the remote and it would turn itself back on!  I finally traced the problem to a ceramic disc capacitor that had gone leaky.  I think it showed something like 100K resistance, but it varied with time and polarity of the meter!

Ed