Voltage Multipliers (for our tube projects...)

[sajy_ho]

Since we're reviving this thread, it seems...has anyone been able to source the 100uH inductor lately? 

I've been thru my usual haunts (Mouser, Newark etc) can can't seem to locate a suitable replacement - the listed parts appear to be out of stock permanently...

Another option would be to wind your own, it takes less than 20 minites to wind, You only need about 10 ft enamel wire and a ferrite rod core...

[GibsonGM]

Since we're reviving this thread, it seems...has anyone been able to source the 100uH inductor lately? 

I've been thru my usual haunts (Mouser, Newark etc) can can't seem to locate a suitable replacement - the listed parts appear to be out of stock permanently...

Another option would be to wind your own, it takes less than 20 minites to wind, You only need about 10 ft enamel wire and a ferrite rod core...

Yeah, but what kind of core, how to figure out what's up with that?  I have ferrite AM ANTENNA core up the ying yang...a couple of power inductor cores, but no way to know what the value of uH would be, how many turns and so on...

[duck_arse]

two transistor LC oscillator - cro/counter - some 5% tolerance caps - paper and pencil - the formulas.

wind on 10, 15, even 20 turns of any wire to hand, hook-up wire even. put it in the osc, read the frequency for each value cap. do the calcs for freq, Al, and inductance. once you have the Al for a particular size/type chunk of ferrite, you can calculate the # of turns for a given inductance.

or, just find the Al, then find the interwebs cheat page for the other numbers. but the long way is much more funn-er.

[amptramp]

No need to wind anything:



This is from:

http://diyaudioprojects.com/mirror/members.aol.com/sbench101/BatteryPoweredAmps/3vHiFi.gif

[GibsonGM]

Yeah, Duck, kind of like I supposed...eventually I will find a source to just purchase a few 100uH inductors, LOL!   Lotsa work, not having many cores to choose from and so on.   I only have 2 of the type I had told you about, which MAY be enough, anyway!

Thanks Tramp, I'm used to/enjoy the high B+ preamps, like the GTFO.  Really curious about all the hubbub around these small power supplies! 

[GibsonGM]

Little more effort yielded:  http://cpc.farnell.com/1/1/18834-murata-power-solutions-1410454-inductor-100uh-5-4a.html

Beefy, 5.4A, just since others have reported too much heat with a 1A inductor...

[Zozobra]

also hello Zozobra, and are you planning for 6V or 12V heated valves?

Hi
I'll probably regulate to 6V for parallel heaters as I have an EF86 that I want to experiment with in the input stage.  I need to dig out Merlins power supply book to look up some regulators.  Hopefully a 7806 will do the job.  I recall there being a bit in the book about using them for low current heater supplies.

[duck_arse]

gibson - I believe there are people that own LCR meters. those people would wind some turns, measure the indutance, calc the Al, then wind the right # turns and remeasure. but I'm quite sure those people don't know half the fun.

Zozobra - I, too, have a (6GW8) 6 volt valve amp planned ....... to build, powered from a 19V laptop supply. and, transmogrifox, one of our members, suggested a circuit that boosts the 19V to 25V, to run the heaters on 6V, based on his high-voltage boost circuit. less waste heat.

this thread:
http://www.diystompboxes.com/smfforum/index.php?topic=112122.0

[Zozobra]

Thanks.  I'll have a look at that.  After years of doing turret builds I think it's time I gave designing a PCB a shot.  I have access to a desktop miller at my local fablab and I'll need to go there to fab the chassis and sleeve so I have no excuses now really.

[GibsonGM]

Let's revive it again    I built a typical Nixie supply from one of the usual schemas floating around.   

What I found is consistent with others' results...you need to turn the adj pot down until you get minimum hum.  Now, using a manufactured inductor, I can get 220V or so before the hum is too great.   Using one I rolled on my own, I can only do mid-100's before the hum is bad.   This suggests the "Q" or some other property of the L is in play.  So, buying one is probably best, or becoming REALLY GOOD at rolling your own, ha ha.

I still want to filter this thru an R-C network, maybe 2.2k with 22u to ground, 450V rating, to remove the rest of the 120Hz hum.  Already tried this, it seems to work, and I am still maintaining about 210V with a 24k load (for just under 10mA, way more than needed).

[amptramp]

You could always use a cold-cathode shunt voltage regulator tube:



The pink ones are argon-filled and the orange ones are hydrogen-filled.  These tubes are used as shunt regulators and the only differences in using them as opposed to zener diodes is that they strike an arc at a voltage a little above the running voltage, they should not have any capacitance directly across them or they will become relaxation oscillators and they have minimum as well as maximum currents.  I acquired 15 0B2's recently in a box lot and intend to use them in some receivers and FM tuners to stabilize operating voltages.  You can get 7-pin miniature or octal tubes and voltages of 75, 90, 105 and 150 volts.  The 7-pin tubes regulate between 5 and 30 mA and the octal tubes regulate between 5 and 40 mA.  They can reduce hum to a value below other noise levels and you could use two 105 volt tubes in series to get 210 volts.

[GibsonGM]

Cool!   Altho it's another heater to power.  I'd like to stick to R-C filtering, if it's possible.  But thanks for that, totally another avenue I hadn't considered! Yesterday's voltage regulators   

[amptramp]

Cool!   Altho it's another heater to power.  I'd like to stick to R-C filtering, if it's possible.  But thanks for that, totally another avenue I hadn't considered! Yesterday's voltage regulators 

These are cold-cathode tubes - no heater at all.  I like replacing tube rectifiers with silicon diodes then using the empty socket for a regulator so the voltage is limited as soon as the unit is switched on.  Just give each tube 10 volts more than the running voltage to strike, don't add capacitance to bypass them and set the resistor values to stay above 5 mA and below the maximum current and you can kill the hum and keep the voltage within a few volts from minimum to maximum current.  Check the 0B2 performance here:

http://tdsl.duncanamps.com/show.php?des=0B2

[merlinb]

Thought I'd point out that these boost regulators are now commonly available on eBay for pennies. Described as:
"8~32V to 45~390V DC-DC Boost Converter ZVS Step-up Booster Module Supply"
Although it says it's good for 40W output I have found that 25W is a more reliable maximum for continuous operation. Above this the transformer gets hot.

[printer2]

I bought those boosters but have not used them with any large load. I have bought universal switching wallwarts for low voltage and found generally they do not do the rated current on 120V but will do it on 230V. I am hoping to use the 390V booster for a 6V6 amp with 12V tubes and a brick PS. The booster is not isolated but the 12V brick takes care of that.

[vigilante397]

How does the ZVS supply do as far as noise goes? Are you able to get fairly clean power from it? I've never worked with ZVS's before, but for $5 it may be worth a try.

[GibsonGM]

How does the ZVS supply do as far as noise goes? Are you able to get fairly clean power from it? I've never worked with ZVS's before, but for $5 it may be worth a try.

I'd assume that, because the switching speeds are so fast, there isn't a whole lot of noise associated with them.   You'd plan some filtering into your design, anyway, so they GOTTA be use-able.    If the one I haphazardly BUILT is so quiet (and it is),  I'd be more than willing to give one of those a shot!

[Transmogrifox]

100uH isn't a magic number.  If making a boost converter of some sort you can probably work with anything between 10uH and 1mH, and then you change switching frequency to make it work where it's happy.

LTSpice is a good tool for working out the best switching frequency -- and there's nothing wrong with operating in continuous conduction mode as long as you have a beefy enough MOSFET to hand the switching losses.

[johngovan1234]

I'm making this alive again. I' sorry. I built this smps but thr voltage keeps on faling on some time then rising again. I dont know why.   it starts with 270 ish. Ofcourse when the tube heats up properly,  it will slowly drop the voltage of the plate. So it goes down and down. But when it hits  some voltage. For example from 270 to 180v then it will rise up again. What could be the problem.

[vigilante397]

What could be the problem.

Sooooo many things Can we see what schematic/layout you used for your build? Perhaps some pics of the build? The only times I've had roaming voltages I wasn't able to control on an SMPS build was because I left out a component or two, which is unfortunately common because with tube builds I always stuff the power supply first and leave everything else off until I can get power right.