[neonixie-l] Re: 'Omnixie' Smart Wifi Nixie Clock released

[Tomasz Kowalczyk]

Thanks for suggestions. The smaller one looks like like it should do the 
job with correctly designed converter - the one I found doesn't look like 
it can dissipate more than 100mW without damage. 
Same story for me and CJ5143 - I've received two of those as free samples.
My converter with CJ5143 wasn't a success and I don't know why yet. PCB 
came out really small (22mm x 15,5mm), the transformer is used as a coupled 
inductor, the driver is MCP1632. After loading it with 2mA, the output 
dropped by 10V and efficiency was under 40%. I guess I need to learn more. 
Now I can only suspect that 300kHz is probably too high for a transformer 
with probably significant parasitic capacitances in it. I'm stil more a 
"trial and error" engineer.

Meanwhile, I came to realisation that LLC converter would be a good choice 
only if I made a custom transformer - it is much more winding ratio 
dependent than regular flyback/coupled inductor converter. 


W dniu niedziela, 18 marca 2018 09:36:37 UTC+1 użytkownik Paul Andrews 
napisał:
>
> Hi Tomasz,
>
> I used this transformer from Wurth in one version of my power supply: 
> http://katalog.we-online.com/pbs/datasheet/74488540250.pdf. It is 4mm 
> high. That version of the power supply will easily drive two tubes from a 
> LiPo battery. It will manage up to four IN-12.
>
> The whole project is documented here: 
> https://hackaday.io/project/27899-nixie-tube-power-supply
>
> The transformer I have settled on for my own purposes is this one: 
> http://katalog.we-online.com/pbs/datasheet/749196141.pdf because it has 
> more headroom than the other one. Of course, it is also physically bigger.
>
>

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/972db543-47f3-453e-bbbf-7b061562b57f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: 'Omnixie' Smart Wifi Nixie Clock released

[gregebert]

Is your transformer (inductor) getting too warm ? It's possible that the 
peak current is larger than Isat; at saturation the magnetic efficiency 
drops way off.

It's not easy to measure the inductor current, because parasitics are 
introduced by the scope-probe. Most often, the ground lead picks-up noise 
and makes your signal look far worse than it really is. I remove the ground 
clip and the plastic shroud around the probe tip. Then, I wrap a few turns 
of bare wire around the exposed ground of the probe (approx 1-2 mm from the 
probe tip) and solder that to the closest GND connection. This minimizes 
the coupling of noise to the scope's ground; I've actually removed several 
volts worth of noise using this technique. This will allow you to view 
clean *voltage* waveforms.

If you dont have an expensive current probe (I certainly dont), you will 
need to insert a small resistance (100 milliohms) and measure the voltage 
across it to infer current. Make sure it is NOT a wirewound resistor, which 
is inductive.

The current waveform should look triangular (increase linearly from 0 to 
peak when the switch is on, then decrease to 0 rapidly when the switch 
turns off). Once you have that you can start debugging where things are 
going wrong.


Converters always seem to run beautifully with no-load; when you start 
loading them down, all kinds of things happen.

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/c37e71e9-ee0a-4da5-9894-311b19ff1fb6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.