Inquiry - Nixie IN-12 used for automotive "gauges"??

5 months 2 weeks ago #11826 by AZ_Ron
Hi Ty. That interior is very similar.
I'm thinking In-Dash, using IN-12 tubes. They mount to the rear instead of on the bottom like IN-14's.
Your help is mucho appreciated!!

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

5 months 2 weeks ago - 5 months 2 weeks ago #11830 by Ty_Eeberfest
Let's hear what some other forum members think about this. If my ideas suck tell me why! Propose something better / simpler / cleaner??

I see a couple different ways to approach this. Microcontroller programming will be required regardless of approach - that's inevitable unless you can find ready to run gauge kits somewhere that I'm not aware of.

Easy approach: well relatively easy. Buy one clock kit with IN-12 display board for each gauge. This gets you the HV power supply, display drivers etc. already designed and proven. Mount only as many tubes as needed on the display boards. Get rid of most of the program, keeping HV and display related parts, write the code to implement the gauge. Repurpose an unused I/O pin (there will be several since you're not using the clock functionality) to bring in 0-5 volt signal from gauge sender. If the senders are 0-12 volt that's okay, a simple voltage divider made from 2 precision resistors will take care of it.

The main drawback I see with this method is you end up with a bunch of circuit boards - one for each gauge - that have to go somewhere. The displays can be remote from the brain boards but the interconnect is about 15 wires per gauge. A cable can be used instead of 15 loose wires so it's not all that bad. The upside is that much of the design work is already done and proven.

More challenging approach: this is what I'd do because I like doing this kind of stuff. Design a circuit board that holds all the display tubes, all the logic, the HV supply and everything else. One big board that fits into the dash opening. Have the board fabricated (etched and drilled) in China - very easy to do these days. Obtain and solder in all the parts myself. I'd base it on a single microcontroller, probably an Atmega328P since I know that chip pretty well. I'd program the micro to implement all the gauges. Same comments as above about senders apply.

The main drawback is that there would be a fair amount of design time involved and there's no guarantee that the first prototype will work. May take a few iterations (on the bench, not in the car) to get the board perfect. There's some reinventing of wheels involved: HV supply, etc. But the upside is that the end result is much cleaner than the first method. There would only need to be +12, ground, and sender wires running to the board.

I haven't mentioned the speedometer. GPS versus a sensor - pros and cons to both.

GPS: no connection to the car except power. Accurate. Not affected by changing rim/tire size. May need an external antenna and I'm not sure you'd like a shark fin on the roof. Loses signal (fails) in tunnels and underground garages. Probably needs to draw some continuous standby power to avoid lengthy delays (getting a "fix") when you start the car.

Sensor: probably a rotary pulse generator aka encoder. Needs connection to the car (use existing speedo gear take-off found at tail shaft of most trannies?) 3-wire cable. Works wherever you go. Works immediately without needing standby power. Requires calibration if you change rim/tire size.

Personally I'd probably go with the speedo sensor over GPS but it's a tough call.

Look into it later when the dust is clearing off the crater.
Last edit: 5 months 2 weeks ago by Ty_Eeberfest. Reason: Typo. IN-12 not IN-2!!

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

5 months 2 weeks ago #11831 by nixie-N00b
Hi there,

so i've got also the idea to set a speedometer as nixie-instrument into my beetle this post also catch me.
by following the link of judge i also found this:

may there you can go on?
unfortunately the creator didnt load up some more videos to see the going on of this project or do some further detailed posts ...

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

5 months 2 weeks ago #11832 by Ty_Eeberfest
Hi nixie-N00b,

Thanks. Did you read the comment section of that article. Hilarious (in the OMG are people really this lame??? sense). I guess none of them have ever been shocked by a Nixie power supply. I have. Many times. It hurts a little but it's no disaster and nowhere close to the pain of getting bit by a spark plug wire! Maybe if you placed one hand on ground and the other hand on +170V some real injury would result but you've pretty much gotta be trying to get hurt for that to happen.

Okay rant over.

So he used a Hall Effect (HE) sensor. That makes some sense because technically it's not being connected to the vehicle. Glue a magnet to a rotating member, mount the HE so that it "sees" but doesn't touch the magnet and boom! you've got a non-contact speed pickup.

I was leaning more toward an encoder (ENC) for more resolution. The more pulses per revolution the better, within reason. An ENC could easily give me 1,000 pulses per rev if desired. Controller could just count pulses over a fixed interval to calc speed. With an HE and magnet there's typically one pulse per rev. I noticed the Hackaday guy complained of poor resolution which is not surprising when counting 1 pulse per rev. Probably the solution to that would be to derive speed from the time between pulses instead of counting them. Not much harder than counting assuming the controller has hardware timers.

Anyway... it's something to think about...

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

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

5 months 2 weeks ago #11833 by nixie-N00b
Hi Ty_Eberfest,

finally we hear each other again ... in this post XD

jepp, the idea is creative but i think there are to much probs to handle just without the sesors.
I also gave this idea up cause in the last consequence, as you wrote the Nixie runs with 170V.

In an old car like a rat rod (there more than each other i guess) all shakes and rattles, the nixies won't like this i guess and the connections will be gone after a while or the numberwires broke by vibrations while driving.
(specially in the beetle you got the further prob the gas tank sits in front of the car, then comes the steering wheel with all the small lights ("Mäusekino") also the tachometer. just in common use a perfect rolling bomb without the nixies to ingite the gas when you do a crash. a rat rot may got not this order of components but i bet the gas will also brun when ignite by sparks ;) )

I also found some years before some burned cables in the car ... after investigate this I found out what happens: the isolation was rubbed of by vibrations while driving, +12V was going though the car clock directly to the ground ... all 42Ah at once. it was pure luck the didn't burn to ash and scrap.
so ... high voltage and cars is may not always the best couple.

For the TS, i mean, a nixie-speedo wouuld be great when its a show'n'shine-car and its rolling on a trailer fron convention to convention

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

5 months 2 weeks ago #11834 by Ty_Eeberfest
I don't think the Nixies with their little 170V 25mA power supply are any more dangerous than the 12V car battery that can supply 300A - 400A of short circuit current.I do think the argument that vibration will kill the tubes might be valid. Might be. I'm not aware of anyone ever having tested this.

Ron said he's building a 1941 Ford truck, so the engine will be in front. The original fuel tank is either underneath the back of the truck or behind the seats - I'm not sure which. I'm guessing he will put in a newer safer "racing type" fuel tank somewhere in the back but I'm not sure of this.

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: 0.101 seconds
Go to top
JSN Boot template designed by