Tubes not doing anything - building the modular IN-14 nixie clock kit

More
10 months 2 weeks ago #8087 by basjetimmer
Hi! I am building the modular IN-14 nixie clock, but now, after the step where the tubes are soldered to the tube holders, I am running into a little problem. When I connect the power, none of the tubes do anything! All six of them just sit there, completely dark. The RGB leds do go through their startup cycle, and the power led lights up. Also, I measure 5V between the VCC and GND, and about 190V at 170V and GND (which might be a bit high?).

So, I see three possibilities:
- I soldered all six tubes so poorly that they don't work (but just by looking at the soldering I find this unlikely)
- All my tubes are broken. This is of course possible, I got them for cheap off of ebay. Is there a simple way to test them? Is it safe to hook the anode and one other pin of one tube up to the GND and 170V test points on the board when powered on?
- I have made a mistake on the board and power is not being delivered to the tubes. Can I test this? By sticking a multimeter inside the tubeholder sockets? And what voltage should I read in which sockets exactly?

Any other thoughts?

thanks!
Bas Timmer

PS. The manual so far was very nice and detailed. I noticed one very small mistake. On page 14, there is a picture of the board after placing the high voltage circuit that seems to be missing R18.
PPS. I have only the tiniest bit of knowledge on electronics and circuits. So please do not hesitate to explain things to me as if i were a small child. And forgive any noob questions (more are probably coming when I finish the build).

Also, in case I am simply missing some required components at this point, a picture:
Attachments:

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

More
10 months 2 weeks ago #8096 by Ian
It's much simpler than that!

You don't have the driver circuitry installed. Please continue with the construction, all will become clear in the next two steps.

BTW thanks for the error report. I'm redoing everything at the moment, including video manuals... ;)

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

More
10 months 1 week ago #8101 by basjetimmer
Thanks! Now I feel stupid, I will just continue the build and hope the tubes will come to life soon!

Just to note, I was following the manual and when I reached page 20, it said:

Once the six tube are mounted on their boards, and you have double checked
the orientation of the boards, hook up the power, and check that the tubes
cycle through their test routine.
The digits “0” through “9” should be shown on each tube.


So, please let me know if I skipped some steps without noticing, but otherwise I think this test should be moved to a little later in the manual.

Thanks again!

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

More
10 months 1 week ago #8102 by basjetimmer
Sorry for all the posts, I think I even double posted earlier, I hope the moderation will prevent that from actually appearing in this thread.

I have continued the build up until page page 23. So, the anode control circuit and the cathode control circuit are now built. I can not yet continue to the separator leds/neon section, because I am missing the required 4.7K resistor (quite possibly lost it myself, no problem, I'll pick one up tomorrow if I have time).

The behavior of the tubes is still the same however. That is, I have not seen any of them light up at all. Is this still expected behavior? Are the separator leds required for the tubes to start their test cycle?

Thanks!

Attachments:

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

More
10 months 5 days ago #8128 by Ian
Hmmm, now it looks complete, and it is not normal that nothing appears. The separators are not required to make the clock work.

I would not expect that you got a whole batch of bad tubes. Even cheap ones are usually OK, but I suppose it could happen that all of them are bad.

OK, here are some things to check:

Once again, measure the HV voltage at the test point. Anything above 160V should be OK. Lower voltages will cause the digits to light, but be "fuzzy".

An easy way to check the tubes is to hold them close to a plasma globe (if you have one). The gas in the tubes should light up. If you don't have a plasma globe, then

Perform a factory reset, so that the colours of the LEDs are cycling. Once this is happening, all of the "complicated" HV handling is disabled (e.g. LDR dimming), and you should get a steady 178-180V at the test point.

Normally the soldering on the tube holder boards is not very critical. I would not worry about it.

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

More
9 months 1 week ago #8166 by basjetimmer
Hi Ian, thanks for your response. About a week after your reply, I went to power up the clock again to shoot a little video of the status (because the LEDS were already cycling through the colours), but to my great surprise it suddenly worked!

I have no idea what happened, I hadn't touched the clock since my previous message here, but everything seems to be fine now so I'm very happy.

I was thinking about getting the WiFi module next. If I purchase the 'WiFi time provider add on' as it is sold on the nixieclock.biz store, will it contain the same parts as if I had purchased the full WiFi kit from the beginning (that is, will I be able to mount it on my modular board)?

Thanks!

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

More
9 months 3 days ago - 9 months 2 days ago #8192 by basjetimmer
Hi!

In addition to the above, I actually have another question: I noticed on some positions the tubes' decimal points light up during some numbers. I have already switched the tubes around, and concluded the effect is linked to the tube position (ie both the seconds-tubes) and not the tube itself.

The one on the tens-of-seconds is always on, the one on the unit-seconds comes on sometime during the display of '2', and turns off sometime during the '6', the exact moment it turns on and off varies, sometimes it's very early during the '2', but then it drifts until it is just right before the '3' appears. I have a little video of it:

giant.gfycat.com/CoolRealisticDutchshepherddog.webm

I was wondering what could be wrong. Should I just cut the leads? Or is there a nicer, cleaner, solution? Or maybe does it indicate some other bigger problem that needs to be solved?

Thanks!
Last Edit: 9 months 2 days ago by Ty_Eeberfest. Reason: Fixed video link (embed didn't work)

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

More
9 months 2 days ago #8194 by Ty_Eeberfest
What happens if you temporarily remove resistor R20? Any change in decimal point behavior?

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

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

More
8 months 4 weeks ago #8209 by basjetimmer
Thanks!
I have removed R20, and it seems to actually fix the entire issue. But I'm afraid to leave the clock plugged in for a long time like this, what is the actual purpose of R20, is it not a necessary component? Is it safe to leave it off, or should I replace it (for a higher valued resistor)?

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

More
8 months 4 weeks ago #8210 by Ty_Eeberfest
Short answer: It's fine to leave R20 off, that will not hurt a thing. You will never again see any decimal points light up, ever, even if you want them to light for some reason in the future.

Long explanation: Look at page 6 of the manual. On the right hand side of the MEGA8-P you will see it's Pin 11 connects thru R20 to the base of transistor Q4. When the MEGA pulls its Pin 11 high, Q4 turns on, in turn lighting a decimal point. Which decimal lights up depends on which tube's anode the multiplexing scheme has turned on at that moment.

By having you remove R20 I "split the problem in half". With R20 removed the MEGA can not light any decimal points because it can't turn Q4 on. If removing R20 had NOT made the problem go away we would know that whatever is wrong is after (to the right of, on the drawing) R20. But removing R20 DID make the problem go away so we know something is going wrong ahead of (left of) R20.

To solve the real problem (rather than working around it by pulling out R20) there are a few things to look at. First look carefully at your soldering quality around Pin 11 of the MEGA. Use a magnifier. Make absolutely sure there are no shorts ("bridges") from Pin 11 to any other pins. Also inspect the copper trace that runs from Pin 11 to R20 for shorts to other traces. The way those decimal points acted kind of random in your video makes me wonder if there's a tiny bit of solder or other metal junk stuck on the bottom of the board near R20 or Pin 11.

If you don't find the problem by inspecting soldering and doing any needed cleanup, then we need Ian to get involved in this thread to tell us what could cause the MEGA to randomly turn on Pin 11 - which I doubt is happening but it's the only other answer if all of the above fail to fix it.

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

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

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.235 seconds

Search

Tube Suppliers

Go to top
JSN Boot template designed by JoomlaShine.com