4-digit IN-18 clock project

[newxito]

After a few months with no time for hobbies, I'm now retired... again…

I've started a new little project, I always wanted to make a 4-digit nixie clock.

I already got the boards; purple is currently my favorite color…

[gregebert]

OSH Park ?  They do excellent work (love that ENIG finish...) but I only use them for very small boards (less than 3 square inches). Even though they are just across town from me, it's faster and cheaper for me to have larger boards fabbed in China (send out Thursday nite, boards delivered on Tuesday, standard thruput time) thru PCBway, and their quality is very good.

Congratulations on the retirement....I'm still counting down.... At least for me it's now measured in years, not decades.

[newxito]

The boards are from China, good quality, fast delivery and good prices. I tried to order in Europe, unfortunately as a hobbyist I cannot afford their prices.

[Richard Scales]

I've had some of the 'purple' boards from JLCPCB - all good - especially delivery to UK.

[newxito]

Today I built the clock. There is an error on the board, the DataIn pin of the LED array is connected to the wrong side of the level shifter, but it seems to work fine with 3.3V. I also had problems with the input protection (over, under and reverse voltage). I always used the circuit with 12V but this boards uses 5V, probably the monitoring window is too small. Anyway, removing some components and a big blob of solder, solved the problem :-)  

I fill fix the design errors on Rev2… time to start making the case…

[Audrey]

Nice work!

On Sun, Nov 7, 2021, 5:14 PM newxito <axt...@gmail.com> wrote:

Today I built the clock. There is an error on the board, the DataIn pin of the LED array is connected to the wrong side of the level shifter, but it seems to work fine with 3.3V. I also had problems with the input protection (over, under and reverse voltage). I always used the circuit with 12V but this boards uses 5V, probably the monitoring window is too small. Anyway, removing some components and a big blob of solder, solved the problem :-)  

I fill fix the design errors on Rev2… time to start making the case…

--
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 view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/6170fd04-351e-476e-9641-13cdd095d1ean%40googlegroups.com.

[ZY]

What is your method of input protection?

[newxito]

This is the protection circuit I use, taken from the LTC4365 datasheet. I used the formulas to calculate the resistors and the LTspice simulation confirmed the values. The circuit works, kind of, shuts down under 4V and over 6V. But when starting with 5V I only get around 4.2 output voltage, that’s wrong. Probably some kind of oscillation…

I use the same circuit with other resistor values for 12V and it works perfectly all the time.

[newxito]

Maybe 5V is not enough to fully turn-on the mosfet, I have to check that...

[Paul Andrews]

Looks really nice

[ZY]

Oh nice, I've been using the same chip but I wanted to double check if there wasn't some other solution people were using.

I went with the SiSB46DN suggested by the datasheet and I've been fine with protection even for 3.3V with a lower voltage bound of 3V.

I've also tested a circuit for 0.6-1.5V for filaments by powering the IC at a higher voltage but applying the lower V-filament to the transistor, and that also works.

[newxito]

Good to know that it also should work with lower voltages.

I remember using the SiSB46DN too. I switched to a MOSFET with SOIC-8 package because for me it's easier to solder. But I paid too little attention to the Vgs(th) value.

[John Rehwinkel]

> Good to know that it also should work with lower voltages.
> I remember using the SiSB46DN too. I switched to a MOSFET with SOIC-8 package because for me it's easier to solder. But I paid too little attention to the Vgs(th) value.

In my experience, people tend to pay too much attention to the threshold voltage. They don't turn on all at once, they're analog devices, the threshold voltage is just where they begin to turn on. For that transistor, it's where it can conduct 250µA. To figure out how much gate voltage you need for a given conductance/current, look to the "Transfer Characteristics" or "On Resistance vs Gate Voltage" graphs. Happily, this transistor conducts well with 3.3V on the gate, but still has some resistance (and therefore dissipation). It'll go down to 10 milliohms, but takes 10V on the gate to do so.

- John

[newxito]

John, thanks for your input. Next time, before I order a MOSFET I will have a closer look to this graphs...

That’s the nearly finished clock. I’m waiting for the MOSFET that should work with 5V. The shipment is still on hold in Dallas… since Monday…