Driver chips besides 7441 or 74141

[John Snow]

What driver chips have you used besides 7441 or 74141 in direct drive mode for your Nixie projects?

I've used Max69** chips with VFD projects but have not found a good Nixie driver chip yet.

[gregebert]

It's like asking an audiophile if they prefer amplifiers that use tubes vs MOSFETs vs bipolar transistors.

You will see compelling reasons for any choice

That said, my preferences:

For true nixies (individual cathode for each numeral) HV5530 with an anode-current regulator

For segmented nixies like b7971, use a shift register such as 74xx595 to control high-voltage  NPN current-regulators on each segment.

[David Forbes]

I use the TD62083 in my Nixie watch. The TD62084 is the 5V input version. It needs a 50V Zener diode on the common diode pin. 

On Feb 26, 2018 7:49 AM, "John Snow" <jrope...@gmail.com> wrote:

What driver chips have you used besides 7441 or 74141 in direct drive mode for your Nixie projects?

I've used Max69** chips with VFD projects but have not found a good Nixie driver chip yet.

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/43b88597-fbb6-4d3f-8af7-c5a5201f9202%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[John Snow]

The watch tubes have a much lower voltage requirement? I thought ~170V was the usual level.

On Monday, 26 February 2018 18:55:31 UTC, nixiebunny wrote:

I use the TD62083 in my Nixie watch. The TD62084 is the 5V input version. It needs a 50V Zener diode on the common diode pin. 

On Feb 26, 2018 7:49 AM, "John Snow" <jrope...@gmail.com> wrote:

What driver chips have you used besides 7441 or 74141 in direct drive mode for your Nixie projects?

I've used Max69** chips with VFD projects but have not found a good Nixie driver chip yet.

--
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.

[Paul Andrews]

I use the HV5523 because it is a 5V chip (though I would prefer 3.3V) and because it has a high sink current.

[gregebert]

Even though the nixies operate around 170V, they provide some voltage drop so that the driver doesn't see the full 170V. If you use a driver rated at a lower voltage, you want to provide protection as Mr. Forbes does (the zener diode) so that the driver doesn't get stressed from leakage current.

The other option is to use a driver rated for the anode supply voltage.

If you do neither of the above, it's possible for leakage current thru the nixie to result in voltage higher than the driver is rated for. At this point, welcome to the jungle.....

Whether or not a driver degrades over time is a matter of device physics.

My own belief is that MOS devices will get destroyed because oxide destruction is voltage-related.

Bipolar (ie NPN) destruction results from excess current (basically melting) so I think as long as the current is very small, on the order of micro-amps, the driver wont get destroyed. But I do wonder if there is long-term reliability impact (degradation) from some other mechanism that would cause leakage to increase over time.

[Jonathan Perkins]

This is placing a zener diode between power and ground to conduct above a trigger voltage?

--
You received this message because you are subscribed to a topic in the Google Groups "neonixie-l" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/neonixie-l/-5Jt1SSYl3s/unsubscribe.
To unsubscribe from this group and all its topics, send an email to neonixie-l+unsubscribe@googlegroups.com.

To post to this group, send email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/7d67a2f9-f36e-45e6-86b8-9feee92fbc93%40googlegroups.com.

[Dekatron42]

The SN75468 or SN75469 are two of my favourites, hooking up the COM pin to some 60V via a resistor divider or zener and resistor works beautifully. I've seen it used with Burroughs Nixies and Burroughs Bar Graph displays. It has been used by a bunch of people here too.

/Martin

[Terry S]

Makes a good relay driver as well.

[Terry S]

Also the ULN2003. Similar part. Some of these older technology parts are getting harder to find.

Terry

[John Snow]

I found a pretty good comparison on http://swissnixie.com/index.php?go=cathodedriving for those interested - I'll be going with the HV chips, just need to narrow it down to which one(s).

[John Snow]

I've narrowed it down to either HV5522 or HV5530 - I can't tell a difference except the sustained/transient voltage for *22 is 220V/230V and the sustained/transient voltage for *30 is 300V/320V.

How do these two differ? I've been through the datasheets for both.

[gregebert]

The difference is either in the overall design of the output driver, and they are specifically designed to handle a max voltage, or they are bin-split during final test depending upon measurements.

Without knowing the device construction (bipolar vs NMOS) it's hard to know if you can push the envelope, so play it safe and stay within spec limits.

If your anode voltage is below 220 volts, the HV5522 should suffice.

I  used the HV5530 with an anode supply of +200V, just for extra margin on my IN-18 clock.

[Mike Playle]

On Monday, February 26, 2018 at 2:49:00 PM UTC, John Snow wrote:

What driver chips have you used besides 7441 or 74141 in direct drive mode for your Nixie projects?

I've used Max69** chips with VFD projects but have not found a good Nixie driver chip yet.

 I've just hooked up a MAX6921 to an IN-12A nixie on a breadboard. From

what I can see, it works just fine, even with all the cathodes turned

off at once.

Admittedly the 76 volt maximum supply voltage seems at first to be too

low for a nixie, but as gregebert says, the full anode voltage doesn't

necessarily appear across the driver, and in practice 70 volts seems

to be enough to switch the tube completely off. Sinking current back

into the rail doesn't seem to hurt it much either.

Is there anything particular that puts you off using MAX69** parts to

drive nixies? If I've missed something, I'd rather know about it

before I buy PCBs! I won't dismantle this breadboard yet because I

still need to do more work on it, so if there are any measurements you

think I should take, I'd be interested to hear about them.

Mike

[gregebert]

The MAXIM device uses BiCMOS technology, so the I/O pins are probably NPN devices and should be more tolerant of voltages above 76 VDC.

It's best to follow the datasheet for all possible operating conditions, but I think you could use a bleeder resistor across the MAX6921 pins and GND.

First, measure the leakage voltage on all of your nixie tubes with a DMM that has an input impedance of 10Megs. Basically, connect the anode of the nixie tube to the anode supply, and measure the voltage at all 10 cathodes to GND. If it's significantly lower than 76 volts and you get no noticeable glowing on the tube, you can use a 10Meg bleeder resistor.

It's important that you take reading on several tubes. I've seen a lot of variation in leakage characteristics of b7971's.

If you measure significant voltage, say 30 volts or more, you will need a smaller bleeder resistor to protect the MAX device. The problem is that when you reduce the bleeder resistor, the tube will glow more and it could be a problem with all 10 cathodes pulled-down.

If you get intolerable glowing with a bleeder resistor that gets you a safe voltage for the MAX device, you should use a different driver.

[Mike Playle]

On Monday, March 5, 2018 at 7:14:56 AM UTC, gregebert wrote:

> The MAXIM device uses BiCMOS technology, so the I/O pins are

> probably NPN devices and should be more tolerant of voltages above

> 76 VDC.

I'd prefer not to allow the voltage to rise that far. 70 volts seems

to be enough to reliably blank all the cathodes on the couple of tubes

I tested.

The data sheet I have says the output stage consists of a push-pull

pair of N and P channel FETs. A switched-on FET will conduct current

in both directions; even a switched-off FET has a body diode that will

start to conduct 700mV or so above the rail. There may well be

protection diodes in there too.

So I expect the chip to be capable of sinking a few uA of current back

from its outputs into the positive rail, and testing seems to confirm

that this works.

> It's best to follow the datasheet for all possible operating

> conditions, but I think you could use a bleeder resistor across the

> MAX6921 pins and GND.

I'm honestly not convinced that this is necessary. I think the driver

can handle the leakage on its own while staying within its specified

operating range. 

> First, measure the leakage voltage on all of your nixie tubes with a

> DMM that has an input impedance of 10Megs. Basically, connect the

> anode of the nixie tube to the anode supply, and measure the voltage

> at all 10 cathodes to GND. If it's significantly lower than 76 volts

> and you get no noticeable glowing on the tube, you can use a 10Meg

> bleeder resistor.

A 10 meg meter reads ~60V (so it's drawing 6uA) and just starts to

light the tube ever so slightly. The leakage current at 70V will

presumably be significantly less, though I don't have the equipment to

measure it easily. I don't think this is going to stress the driver

too much.

(This does prove that I have to be careful using my multimeter to

measure these voltages, though!)

Mike

[gregebert]

Like the other thread about the FLW clock, I missed the fact that this driver also has rail-to-rail drive with a high-voltage supply pin (76 volts).

That will clamp any leakage current from the nixie tubes, so a bleeder is not needed. ESD structures in the IC are designed to clamp hundreds of milliamps (for a brief time), so a few uA will not do any harm.

What was your anode supply voltage when you measured the leakage of your nixies ? That will give us an idea how much of a voltage drop your nixie has when it's just starting to ionize. My rough guess is your tube will have at least 125 volts across it as it just starts to faintly glow. If you clamp your driver at 70 volts, you should be able to run your anodes up to +195V, which is more than enough. 170 to 180 is typical.

As you alluded to, your actual leakage current thru the driver chip  might be far less than the 6uA you measured with your DMM.