Problem with Arduino Nixie Clock

More
2 months 3 days ago #12721 by Ty_Eeberfest
I've been waiting hoping Ian would have some idea what's going on with your clock. I am out of ideas, at least for the moment. I see nothing wrong with your inductor and MOSFET replacement choices.

In the interest of completeness... have you done the factory reset + HV calibration routine after every component change? It's important you do so (or at least try - not sure if you can make it to the HV calibration step without burning up).

Factory reset = hold down button as power is applied

HV calibration = while the tubes are counting after the reset, push the button momentarily when the tubes are showing 888888. This should initiate a ~30 second routine of counting up & down, flashing LEDs etc. This is "tuning" the characteristics of the PWM gate drive to suit your power supply and HV parts. Normally when the routine is finished the clock goes into normal timekeeping mode.

All of the above is probably "review" for you but I just want to be thorough!

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
2 months 2 days ago #12722 by KRiZ
I have been doing a factory reset every time. When I do that, 4/5 tubes will start counting. But after a few seconds, magic smoke begins to appear from the Mosfet. Sometimes I get far enough to push at 888888, but it makes no difference.

Please Log in or Create an account to join the conversation.

More
2 months 2 days ago #12723 by Ty_Eeberfest
Okay you're doing it right then. Thought so but wanted to be sure. I have a couple farfetched ideas for you to try...

1 - I don't think I've asked this: with power on, hopefully before the smoking begins, what DC voltage do you read at Arduino A0 pin (the HVS signal)? About 2V is expected. If the reading is drastically off we have something else to look at - namely the voltage divider formed by R1 & R2.

2 - With power disconnected for a while (so caps can discharge) read resistance from +170 test point (+ probe) to ground test point (- probe). This isn't a precise measurement because there is a reservoir cap across +170 & ground, but a rough reading will do. Should initially be at least 100K, probably higher, and will change as the capacitor charges off your meter. If the reading is low we at least have something else to look at.

3 - Try eliminating the "consumers" of 170V. Pop the 6 opto-isolators out of their sockets. Since you are using neon separators, you should also remove the jumper at P1 that selects between VIN and 170 for the separators. These steps will remove all legitimate loads from the 170V rail. Power up. Smoke? +170 test point reading >= 170V? If no smoke and voltage good you can start putting things back and see what makes the problems reappear.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
2 months 18 hours ago #12724 by KRiZ
The Arduino A0 pin measures about 2.19V

The resistance between 170V and ground are about 360KOhm

I have eliminated the 6 opto-isolators and the jumper. The mosfet and inductor stayed cool, so no smoke :-) The voltage was about 182V

After that I started to put the opto-isolators back one by one. The bulbs started to light. At last I resoldered the bridge and it looked like the clock was working again.

But the clock was very dim, so I reattached the wifi board to disable the ldr and set the dim level back to a normal brightness. Then the inductor was getting hot.
It was getting too hot to touch and I could start smelling it, so I turned off the power before it burned completely. The mosfet stayed cool all the time.

Please Log in or Create an account to join the conversation.

More
2 months 13 hours ago #12725 by BS_Jim
Does anyone else think that this inductor looks like it's too lightweight for the job?



I just bought a bunch of the ones like Ian sends with the kits, from ebay, for use in the clocks I've built. They weren't expensive, they can handle an amp or more, and I haven't had an issue.

Please Log in or Create an account to join the conversation.

More
2 months 9 hours ago #12726 by Ian
Hi Jim, what does the spec say on it. It looks a bit flimsy, but who knows what's going on inside it...

Please Log in or Create an account to join the conversation.

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.100 seconds
Go to top
JSN Boot template designed by JoomlaShine.com