Pete Hand FLW Clock issues

[Richard Scales]

I am just looking at an FW4-RI rev B board from Pete Hand. It as clearly had some liquid spilt on the tube side of the PCB and was only lighting the two tubes on the RHS and then in a very odd manner.

After a good scrub with some Isopropyl alcohol I started to investigate and, thanks to the relative simplicity of the multiplexed design I was soon able to deduce two obvious faults.

One being that there was a break in the PCB trace for one of the cathodes. The other being that there was a break in the PCB trace that connected the HV from one set of high-sided drivers to the other.

I 'fixed' these two breaks using wire links for now and that certainly restored the operation of the two previously non functional tubes.

The remaining problem is that all segments on all tubes are illuminated dimly whilst it is  also possible to discern correct word display on all 4 tubes if you really concentrate.

All cathodes have between 80K and 90K resistance to ground and I am assuming that this is the reason for the partial illumination of all segments. Though I also note that all the traces from the cathodes to the HV driver via the relevant series resistors is on the clean/underside of the PCB where I see no evidence of any damage.
My questions (at this stage!) are based on my assumption that there should be an infinite resistance between un-used cathodes and ground. I have never used an HV5812 though am familiar with the HV55XX and HV56XX range.

Is this a 'common' fault on these boards?
Is this a feature of the HV5812 (I am guessing NOT)?
As the resistance between all cathodes and ground is between 80K and 90K - I am  assuming that this is a fault at the driver level and not some more failed traces between the cathodes on all tubes.

What kind of interest is there in these boards?

All pointers gleefully received!

- Richard

[Richard Scales]

This shows the area where there is a break in the HV supply between the two sides of the board

There was no connection between the two pins ringed in yellow.

I think this is displaying TE?E though I could be very wrong?

- Richard

[Richard Scales]

Update:

The two major faults were that the left hand two tubes did not light and one segment was 'missing' on all tubes. 

Both these issues were 'fixed' by circumventing failed PCB traces with wire.

Then the secondary issue of all segments lighting was ultimately resolved by replacing the HV5812 driver.

Now I have a working Pete Hand 4LW clock in  case!

 - Richard

[gregebert]

I looked at the HV5812 datasheet, and I see that it is only rated for use as a VFD driver up to 80 volts. As long as the VPP pin is properly clamped so that it wont exceed 80 volts it should be OK to use it for nixie tubes, but you will need to make sure the anode supply isn't too high, otherwise you run the risk of ghosting or faint glowing of the cathodes when they should be off. I'll spare everyone of my rant against multiplexing when direct-drive is possible.

It is very possible to measure finite resistance from cathodes to ground because the HV5812 has diode-clamped outputs. What you are measuring is the resistance thru that diode, into the VPP supply, and from there to ground. Most likely there is a bleeder resistor for the +80V (or whatever Pete uses) to GND.

[Richard Scales]

Interesting - and helpful - perhaps there was some other issue with it but replacing it fixed the 'all segments illuminated' issue - it's now been running non stop for a couple of days - all good.

So suddenly I have three 4LW clocks, one by Raymond Weisling, one by Pete Hand and one be me!!

Whilst I appreciate there is lots of history behind those first two - I already have too many clocks and wonder if anyone out there has a need that is greater than mine - I am open to offers on both the Raymond and Pete's clocks - both without tubes.

Please send PM if interested.

- Richard

[Richard Scales]

@gregebert - My FLW design is direct drive and utilises the original 'Cinch' sockets to hold the tubes.

- Richard