Chronotronix V400 clock dead power supply

[David Forbes]

Folks,

I was recently mailed a Chronotronix V400 clock to see if I could repair
it. This is a 6 digit clock with bi-quinary B-5025 tubes and a MAX1771
power supply.

It has a blown fuse, and the IRF840 ran hot for a while (the board is
blackened). It and the the HV filter cap were replaced, as the solder
joints are newer. The little L4 inductor next to the IRF840 looks like
its solder joint overheated, as the PC board there is damaged.

The anode resistor on the leftmost digit even anode is also cooked.
Perhaps this tube was bad and overloaded the supply for a while?

Do any of you have experience in working in this clock or similar,
and/or do these symptoms sound familiar?

--
David Forbes, Tucson AZ

[Nick]

You could always just email Dieter!

Anyway, most MAX1771 designs are remarkably similar - whilst failure is rare, its generally down to a few possibilities...

• The MAX1771 failed and is keeping the FET on, which cooks it as the current limit resistor (if used) with be ignored. This will also cook the inductor and short the input low voltage DC supply
• The FET fails with D-S short. Similar symptoms to the first mode
• The IRF840 is not a good choice - high Qc, high RDSon - IRF644 is a far better choice (as are many others). It may be the the IRF840 is not the original FET and was replaced by someone who thought any HV FET would do.
• The supply, if designed and running properly, should not be capable of being overloaded - it has a current limiting resistor.
  
• The inductor has an internal short (rare failure mode).
• The inductor has been replaced with one whose maximum DC current is too low - for most nixie PSUs, at least 1A capacity is required
• For some reason, there is high frequency parasitic oscillation in the PSU - MAX1771 designs normally run at about 60kHz - if the thing has parasitic oscillation (typically in the MHz region), the FET will cook.
• The only other active component, the diode, may also have failed - should be a low Trr (reverse recovery time) and preferably low Qrr (stored gate charge). ES2F is a good one.

Have a look at my MAX1771 page (http://www.desmith.net/NMdS/Electronics/NixiePSU.html) for pretty much all you need to know about them along with recommended components.

EDIT: Small extra possibility - If the HT have been completely shorted, as opposed to being simply overloaded, the DC supply to the SMPS is effectively shorted through the coil & diode, possibly cooking both (and the HT supply). Just a thought.

Nick