Tube Hobby woes

[MichaelB]

Had another TubeHobby clock start blowing fuses the other day and I know others on this forum have had similar issues. I have 4 of these muxed clocks and 2 that drive IN-18’s, the others use smaller tubes like the Z566M and the Z574M and have been fine. However, I have had issues now with both of the IN-18 clocks where the 1^st blew C6 (see attached PS schematic) and now this one has eaten (shorted, or darn near, measures .3ohms) the 270uf inductor.  Replaced it and the clock has run fine now for 3 days, so its fixed for now. Not sure what it is about this clock's power supply driving IN-18’s and so wondered if others have noticed this problem and what a long term cure might be? Included the power supply schematic for review.

[MichaelB]

Oops said uf!! Doh!

[Terry Kennedy]

On Thursday, September 1, 2016 at 1:51:45 PM UTC-4, MichaelB wrote:

Had another TubeHobby clock start blowing fuses the other day and I know others on this forum have had similar issues. I have 4 of these muxed clocks and 2 that drive IN-18’s, the others use smaller tubes like the Z566M and the Z574M and have been fine. However, I have had issues now with both of the IN-18 clocks where the 1^st blew C6 (see attached PS schematic) and now this one has eaten (shorted, or darn near, measures .3ohms) the 270uf inductor.  Replaced it and the clock has run fine now for 3 days, so its fixed for now. Not sure what it is about this clock's power supply driving IN-18’s and so wondered if others have noticed this problem and what a long term cure might be? Included the power supply schematic for review.

That clock is very hard on the HV cap. I've had problems in the HV section take out that cap (and usually the fuse) and sometimes even the wall AC adapter. It happens infrequently enough (in a population of 4 clocks) that I haven't bothered seeing how hard it would be to ditch the built-in HV and use something like a tayloredge packaged supply.

[David Forbes]

Michael,

Those issues are most likely to be caused by poor quality components, rather
than a design flaw. The inductor has a high voltage across it, and the capacitor
handles a large ripple current.

You might want to look into higher-priced parts. Military grade capacitors and
inductors cost less than a single IN-18 tube.

On 9/1/2016 10:51 AM, MichaelB wrote:
> Had another TubeHobby clock start blowing fuses the other day and I know others
> on this forum have had similar issues. I have 4 of these muxed clocks and 2 that
> drive IN-18’s, the others use smaller tubes like the Z566M and the Z574M and
> have been fine. However, I have had issues now with both of the IN-18 clocks

> where the 1^st blew C6 (see attached PS schematic) and now this one has eaten

> (shorted, or darn near, measures .3ohms) the 270uf inductor. Replaced it and
> the clock has run fine now for 3 days, so its fixed for now. Not sure what it is
> about this clock's power supply driving IN-18’s and so wondered if others have
> noticed this problem and what a long term cure might be? Included the power
> supply schematic for review.
>

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to neonixie-l+...@googlegroups.com
> <mailto:neonixie-l+...@googlegroups.com>.
> To post to this group, send email to neoni...@googlegroups.com
> <mailto:neoni...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/ff7e483e-96e6-4172-b9b8-0bae8a326a2b%40googlegroups.com
> <https://groups.google.com/d/msgid/neonixie-l/ff7e483e-96e6-4172-b9b8-0bae8a326a2b%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.


--
David Forbes, Tucson, AZ

[gregebert]

Most likely, the IN-18's are using more current than the supply was designed for.  Switching power supplies work great with zero or minimal loading, but once you start cranking more current thru them all sorts of problems can arise. I've found this out the hard way several times.

Look for signs of trouble, such as overheating (especially in the inductor from saturation), excess ripple voltage on the cap, voltage spikes on the MOSFET, duty-cycle (longer MOSFET "on" time will cause more current in inductor & MOSFET).

[MichaelB]

Tending to think you're right about the load the IN-18s present vs. the other tubes since both of those smaller tube clocks have run for years without an issue. And this last go round I did replace all the electrolytics with Xicon caps and actually used a 350V vs. the required 250V for the HV cap. The inductor was a Bourne and I replaced it with same.
Time will tell, I guess.

[gregebert]

If you have a scope, take a look at the MOSFET 'on' time with 1 tube running vs. 6. From there, you can estimate the peak inductor current (I=Vin*Ton/L) and compare to the datasheet.

[Dekatron42]

Some time ago John Taylor (Tayloredge) converted a wall wart into a Nixie PSU and showed the result here on the forum, after that I started to experiment with other small switching power supplies. In the answers to my questions John mentioned better rectifier diodes and also that he used two in series. I tested two UF4007s in series and found that the switcher worked a lot better, with one it sometimes blew up and caused a lot of collateral damage to the rest of the switcher. At the same time I pulled out a few switchers I had lying in a box like the ones here and I added a second rectifier diode to those too and they also worked a lot better. I can't explain this as I know too little about electronics in this area but I am sure that others can. I also discussed snubber and found that I could skip those if I used two UF4007s in series as it looked good enough on a scope then. The result with two diodes in series was that they didn't blow up but I also got a lot cooler inductors which also led to less heat in the switching transistor so there was a lot less power loss in the switcher. I tested this change on both switcher with a single inductor and also on switcher which uses an isolated transformer. I wish I knew more about this so that I could explain it properly but unfortunately I don't.

Don't forget to replace the mosfet if you have blown the inductor as the mosfet has seen a lot of stress during these situations and is probably partially damaged. I have had to replace the mosfet on several occasions even though it at first seems to work well.

/Martin

[MichaelB]

Thanks for the advice Martin. I may as well change the MOSFET before buttoning this thing back up in it’s case. I have plenty of them. I could certainly add another diode in series if this is the general consensus. I'm really in no hurry to button it back up if it means fixing these recurring issues.

[Nicholas Stock]

Michael, sorry to hear you're having issues. I've built probably more than 30 of these things over the last 10 years for friends and family and have only had one issue so far (bad DS1307), Jonas was very helpful in helping me diagnose. Did you reach out to him? I was never happy with the lack of a good heatsink at the regulator and mosfet (the board plane didn't seem enough to me, but I'm no expert in the least) and always added a small custom heatsink under the 7805...maybe heat is your enemy in a case? Apologies if this seems a trivial point to make!

Cheers,

Nick

--

You received this message because you are subscribed to the Google Groups "neonixie-l" group.

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+unsubscribe@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/961809a3-2d54-490d-a6d9-360d14c9a7a2%40googlegroups.com.

[Terry Kennedy]

On Friday, September 2, 2016 at 12:48:25 PM UTC-4, gregebert wrote:

If you have a scope, take a look at the MOSFET 'on' time with 1 tube running vs. 6. From there, you can estimate the peak inductor current (I=Vin*Ton/L) and compare to the datasheet.

The TubeHobby clock is multiplexed, so only 2 tubes should be on at one time. 

[MichaelB]

No, not trivial Nick. And I do believe it is heat when driving IN-18s. Too much of that nasty stuff is generated for whatever reason and components fail. I have experimented with setting the voltage way down to 160 and added ventilation to the one that is enclosed, but still after time they fail. For whatever reason these electronics are more susceptible to heat than all other muxed designs I have.

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To post to this group, send email to neoni...@googlegroups.com.

[Andy Williams]

I built a TubeHobby clock in 2006. A couple years ago it quit working and kept blowing the fuse. During troubleshooting, I found that the 170 power supply measured 24 ohms to ground. I removed C6 and the 24 ohm resistance went away. C6 checked good, so I reinstalled it and the clock worked. I don’t know where the low resistance came from. Tin whisker? Solder splash? Circuit board manufacturing defect? The clock is still working 2 years later. 

On Sep 1, 2016, at 12:51 PM, MichaelB <badn...@badnixie.com> wrote:

Had another TubeHobby clock start blowing fuses the other day and I know others on this forum have had similar issues. I have 4 of these muxed clocks and 2 that drive IN-18’s, the others use smaller tubes like the Z566M and the Z574M and have been fine. However, I have had issues now with both of the IN-18 clocks where the 1^st blew C6 (see attached PS schematic) and now this one has eaten (shorted, or darn near, measures .3ohms) the 270uf inductor.  Replaced it and the clock has run fine now for 3 days, so its fixed for now. Not sure what it is about this clock's power supply driving IN-18’s and so wondered if others have noticed this problem and what a long term cure might be? Included the power supply schematic for review. 

-- 
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/ff7e483e-96e6-4172-b9b8-0bae8a326a2b%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

<TubeHobby NCV2.1_manual+Schem. 21.pdf>