Nixie supply - inductor rating?

[GibsonGM]

Hey,

I've been checking out Nixie supplies - found the long 'end all' thread on here, but I didn't find a conclusive answer to what power rating and what value inductor should be used for the typical 555-based ones.    2.5A, like this one?  I should mention that of course this is for experimenting with tube preamps.  I don't know if anyone ever figured out if 100uH or 150uH was the magic number, either.

  http://www.mouser.com/ProductDetail/JW-Miller/PM3340-101M-RC/?qs=sGAEpiMZZMsg%252by3WlYCkU%2fg2BVck4dUGc1qJydB3zuk%3d

This is the basic schematic I've been looking at, same as everyone else:  http://www.ledsales.com.au/kits/nixie_supply.pdf

Thanks!

[psychedelicfish]

I'd say a 2.5A inductor is fine for this application.

Back when I had a 6AQ5 champ kind of thing on my breadboard I was powering it with a supply like that, and found that the output sagged quite a bit, so I was trying to squeeze as much power out of it as I could. I had a 5A inductor and was using a 10,000 uF capacitor and a 36V input to try and dump as much current into the inductor as I could, but that was to supply something like 50mA to my circuit. Unless you're doing something like that, a 2.5A inductor should be fine.

Another thing to note is that toroidal inductors don't seem to work well i this application. When I was playing with that supply on the breadboard I found that the inductor got very hot (even before running it with the 10,000uF cap and the 36V input). I've heard of other people that have used toroidal inductors and had problems with heat too. I'm not sure why this is, perhaps one of the experts could explain?

100uH or 150uH will work, 150uH will store more energy but will take longer to charge during each positive cycle of the 555.

[GibsonGM]

Thanks, Psychedelic.  Yeah, I'm not planning on this needing to supply more than 10 to 15 mA WORST case, really.  I'm hoping to set up 2 12AX7s...three gain sections, CF and stack.  We'll see!  But a good project, even if it stinks.    If it can't hack it, I'll do a single tube boost!

)   

[merlinb]

2.5A is ample. I've even used the puny signal inductors, and they didn't eve get warm. 150uH is preferred.

[PRR]

The current goes by the voltage ratio.

If you output 10 times the voltage, you obviously must suck 10 times the current.

If you hope for 170V out, from 9V in, that's near 20:1 ratio.

If you would really need 15mA for 4X 12AX7, 15mA*20= 300mA or 0.3A.

That's perfection. It will really be more. In a big costly project you want a sharp pencil. In small work, just double the number. 0.6A. 0.5A is worth a try.

However-- one unit of 12AX7 in an audio circuit is usually more like 1mA. Using Fender 100K+1K5 bias, with a 300V supply, you will have 90V across the 100K resistor and 0.9mA per unit. At 170V supply more like 0.5mA each. 2mA for two double-bottles. Reflected back to 9V, 40mA. Buy 80mA of coil.

> if 100uH or 150uH was the magic number

That depends totally on frequency. If it is reasonably right, it works out.

[Johan]

And stay away from toroid core as it will saturate and leave you unhappy with Less voltage than you hoped for

[GibsonGM]

The current goes by the voltage ratio.

If you output 10 times the voltage, you obviously must suck 10 times the current.

If you hope for 170V out, from 9V in, that's near 20:1 ratio.

If you would really need 15mA for 4X 12AX7, 15mA*20= 300mA or 0.3A.

That's perfection. It will really be more. In a big costly project you want a sharp pencil. In small work, just double the number. 0.6A. 0.5A is worth a try.

However-- one unit of 12AX7 in an audio circuit is usually more like 1mA. Using Fender 100K+1K5 bias, with a 300V supply, you will have 90V across the 100K resistor and 0.9mA per unit. At 170V supply more like 0.5mA each. 2mA for two double-bottles. Reflected back to 9V, 40mA. Buy 80mA of coil.

> if 100uH or 150uH was the magic number

That depends totally on frequency. If it is reasonably right, it works out.

Thanks, Paul.  I'd run it off a 12V, probably 500mA, wall wart.  Hoping for about 170, ayuh.   If that was too unstable, we can adjust - there are many warts in the box )

-- Jeez, by the sound of it, a Rat Shack 100uH RFC might work?!   They're rated to 2A...for some reason I have a few of those in the box.  Ferrite core, non-toroidal.

[psychedelicfish]

I'd run it off a 12V, probably 500mA, wall wart.  Hoping for about 170, ayuh.   If that was too unstable, we can adjust - there are many warts in the box )

-- Jeez, by the sound of it, a Rat Shack 100uH RFC might work?!   They're rated to 2A...for some reason I have a few of those in the box.  Ferrite core, non-toroidal.

Any voltage from this circuit should be pretty stable, as the output is regulated (voltage divider from the output connected to the base of a transistor which controls the pulse width of the 555's output). As long as you don't pull tonnes of current, which you won't, you'll be fine with 170V. You'll probably be fine with even higher voltages.

[GibsonGM]

Sweet.  Yeah, I'm mainly interested in getting up high enough to not be starved plate, but don't necessarily care about having the full flexibility that a real PT at 300V would offer.   With 3 gain stages, I should be able to get some good natural crunch with a reasonable onset at these voltages, I'd think...

[PRR]

> a Rat Shack 100uH RFC ....Ferrite core, non-toroidal.

Sure that's ferrite? Last I noticed they were crap-iron core. Which is actually better for car ignition hash reduction because it is very lossy. OTOH, in power conversion the losses suck.

WTH, try it. But would be good to meter input and output V and A, do math, see if efficiency is over 50%. You'd like over 70%, over 90% is only in-your-dreams (or after extensive engineering). Maybe no big deal if input power is abundant and nothing runs too hot.

[GibsonGM]

...I'll get a couple of L's, see what does best!    Yep, that's an iron core, easy to see when you take it out of the yellowed packaging.

I bought a bunch of crap years ago on closeout (like everyone), and try to find places to use stuff...probably not here, tho!

[sajy_ho]

Hi, I'm trying to build a 555 smps for my overdrive tube pedal using GTFO's schem:http://dl.dropbox.com/u/15275178/blast%20from%20the%20past/BFTP_SMPS.svg
Now I have the circuit on bradboard and getting about 190V.
The voltage is ideal for my design bt my problem is the MOSFET, it gets hot very fast even without any load.
I found two inductors; one is this type: http://img.dxcdn.com/productimages/sku_269256_2.jpg rated: 100uH - 3A. The other one is a toroidal rated 120uH. The first one doesn't get hot but the second gets really hot. But with both, the mosfet gets hot as hell.
Any idea why?

[goldstache]

What is your supply by chance?

Also, I have been building Tube Mic preamps and pedals, using submini tubes, and I get a lot of high oscillations with my SMPS design.  It's a take on the Frequency Central SMPS.  It's really solid as far as voltage and well filtered, but in the end, I find layout is introducing a ton of noise. It's all high whine from the 555. 
Any suggestions.  Typically two tube designs.
My Supply is 15VDC @ 3A, regulated to 12VDC for heaters (series), smps on 15VDC rail. 
SMPS is putting out 240VDC B+
Plates are seeing about 90VDC V1A and 120VDCV2A
Any suggestions on how to tame this or solve?
Thanks!

[sajy_ho]

Yep, me too, Too much high frequency oscillations. I'm giving up the 555 SMPS, how about using another chip like LT1054 for a SMPS design. Or this one with a MAX1771?http://desmith.net/NMdS/Electronics/NixiePSU.html
Btw I'm using a 12V-3A switching adapter and a 12V-1A regulated one, with both experiencing oscillations and IRF740 gets very hot, but a little less with regulated one. Any reasons?
Thanks

[duck_arse]

are you both using 555, or the cmos type, like 7555 or tlc555 something? the cmos chip draws much less current, and has a different output section which means less power line spikes. you will probably still need to isolate the 7555 supply from the rest of the audio sections, and keep all your earths seperate. and big fat bypass caps, complemented with little skinny ceramic types.

[sajy_ho]

are you both using 555, or the cmos type, like 7555 or tlc555 something? the cmos chip draws much less current, and has a different output section which means less power line spikes. you will probably still need to isolate the 7555 supply from the rest of the audio sections, and keep all your earths seperate. and big fat bypass caps, complemented with little skinny ceramic types.

I'm using a NE555 timer chip, how about the MOSFET?it gets very hot as I turn on the circuit.

[duck_arse]

I'm not familiar w/ the circuit, but the 555 spikes might be overworking the mosfet. IDK. perhaps work out the currents and voltages the mosfet is actually handling, then see what its specs say.

[goldstache]

In all my builds, as long as my inductor amperage rating was big enough, I had very little heat(with sinks on).  I think you mentioned your inductor being high enough. I may need to faraday?

[sajy_ho]

I found it! The problem was that breadboard. Now I'm getting 200-350V without any heat:http://i587.photobucket.com/albums/ss312/sajy_ho/23122011466.jpg
Thanks to GTFO layout...

[GibsonGM]

Nice!  I have to order my parts so I can do this.