First nixie clock build, struggling to set the time - help, please.

More
2 months 2 weeks ago #13150 by Ty_Eeberfest
Quick FYI on power supply requirements...

First, the power needs to be reasonably clean. That is to say, there should not be significant ripple in the DC, and no high frequency crud leaking through from the switchmode operation. This can only really be tested with a scope, but if you change wall warts and problems go away, well, that's a pretty good test too! Most wall wart supplies are clean enough but you do occasionally run into one that's not.

Running "base load" of these clocks is about 250mA - 300mA. That will vary depending on your settings, especially the colored LED settings. The "base load" is what you can see on a meter but there's actually more going on, thus the recommendation for at least a 500mA power supply. The WiFi module is (like anything WiFi) constantly chattering with the access point. Every time it transmits it draws a pulse of current, LEDs are pulsing, the tubes are multiplexing, and so on. This is happening too fast to see much on a meter but the power supply has to be able to supply enough current for the surges. There is a reservoir capacitor on the clock board to help smooth out the power surges but it can't make up for an inadequate supply.

When running from a bench supply it makes sense to set current limiting to 1 amp. That's a limit - the clock will draw the current it needs and no more. Voltage setting is debatable. 9V is the official recommendation. I run mine on 12V with good results (mainly because 12V wall warts are common here but 9V ones not so much). Anything over 12V is pushing your luck.

Lastly, doing a factory reset any time you change power supplies is strongly recommended. Everything that happens during a factory reset, after the step where you press the button when it's displaying 888888, has to do with calibrating the boost converter (180V supply) for optimum performance on the supplied voltage. If the supplied voltage is changed, even a little, the boost converter should be recalibrated. If you don't do a reset it may still work fine but it also may overheat or supply incorrect high voltage.

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 weeks ago #13153 by Swany
Thanks Ty.

A very comprehensive answer.

I have powered the clock from my new 1500mA 9vdc supply, factory reset, have re-flowed a few joints on the PCB, everything is now working again.

This is to be part of a 50th birthday present for my wife.

What I intend to do is build another clock, as a backup. Then if something happens I can just switch the guts from the case.

I also have a 12vdc PSU warts that I can use, seems to be trial and error with these supplies. The first 9vdc 1000mA PSU's I bought from Amazon as a pair, They both may be unsuitable for this type of project.

Thanks.

Lee

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

More
2 months 2 weeks ago #13154 by Walter John
You wife will love the clock. A lot of cheap switching power supply are indeed unsuitable. I am not sure of the following theory, but with a linear 12 volt power supply I never had any power issues (having build and designed > 20 clocks of different designs); perhaps there is an issue when you use an external switching power supply combined with the switching high voltage power supply circuitry within the clock. Again, I am not sure of this theory, I can only rely on my observation. That having said, perhaps your re-soldering did the trick.

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

More
2 months 2 weeks ago #13155 by Swany
Thanks Walter, I appreciate the info from someone that has 'been there'!
I think I will try a decent 12vdc PSU. What I did notice is that I had far fewer issues when I was using my desktop PSU set to 9vdc and regulated to 500mA.

I have another clock to build (thanks Ian), so hoping I can get this up and running.

I feel more confident now to tackle another.

Another set of 6 Nixies en-route :-)

Thanks.

Lee

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

More
2 months 2 weeks ago #13156 by Ty_Eeberfest
Lee, I'm not so sure setting a current limit of 500mA on your bench supply is the best way to go about it. The clock will draw what it needs - no need to limit current so low at the PSU. If you set the limit at 1A you leave plenty of headroom for those brief spikes I was talking about above. Having said that, setting a low current limit when you power up something new for the first time can limit damage if the unexpected happens.

Linear power supplies... yeah they are big and inefficient but they are still the cleanest power supplies IMHO. I have nothing against good switchers but there are so many out there that are not very good.

It seems like there is a race to the bottom going on with the cheap switching PSUs sold at the usual online places. Anything to lower the price by a few pennies even if it means lower quality power or compromised safety. Reduce that component count - that's all that matters! Average buyer probably won't notice because they'll use it for dome less demanding job like charging batteries. And if they do notice or the product burns up on them, they're not going to bother making an issue about a $10 item failing.

I have a few eBay PSUs here that are so minimalist that overcurrent protection has been left out. Short the output? PSU doesn't care, it keeps running until all the smoke pours out of it. I don't know why I was surprised the first time one of them burned since I only paid something like $5 each for them.

Okay, rant concluded.

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 1 week ago #13165 by Swany
Thanks Ty.
You are correct, all these modern supplies are very light and cheap.

Linear supplies are harder to acquire without cost. I have found this Sinclair ZX Spectrum UK1400. I used to have this when it came out. The PSU is very heavy so almost certainly linear, dating from the 80's. Rated at 9vdc and good for 1400mA.

The other thing I noticed with the build and seemed strange at the time and more so now I have encountered the same issue on my second build - R17 and R23 are stated as 4.7kohm resistors, but I only have one in my kit??

I sourced another from my own stock as I thought it was an over sight. However have had to do the same again on this build.

The 4.7kohm resistor is larger than the others so assume 0.25w.

Am I missing something here or just an issue with the supplied kit.

Thanks
Lee

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

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.098 seconds
Go to top
JSN Boot template designed by JoomlaShine.com