Quick and easy Tube Microphone Power supply

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

maxwall

Well-known member
Joined
Nov 17, 2004
Messages
1,134
Easy tube microphone power supply for 12AX7 or 12AT7 tube mics and more. easily modified to other voltages.

-filament linear regulated (78xx), HT Linear regulation using BUX 85 power transistor.
-easy self etch board design
-low cost

Power Transformer :
primary 230 or 115 vac
Secondary 1 12-16 vac .6-1amp,
Secondary 2 120-133vac 20-30ma
Note : Two separate transformers can be used in series. Pri 115 or 220vac :12vac 6va and 9-12vac : 220vac Secondary. 3-6va

construction notes:
BUX 85 NPN power transistors base must be heat sink mounted with insulator and white heat sink compound. Use
of teflon plastic insulated washers recommended to prevent component short to ground.

7812 Filament regulator does not require insulator between the component base and heatsink.

power supply can be modified with a 7806 filament regulator and a change to 12vac secondary power transformer for use with tubes requiring 6 vdc filament voltage. examples are 6922, 6072, 6205, 5840 etc ... depending on application , keep in mind a 6 vdc filament may require double the current (.300ma min ) of the 12vdc filament (.150 ma ). consult your tube datasheet for filament requirements.

1N5380B - 5W , 120v zener , (B) 5% tolerance - Mount this zener so its spaced above board for cooling.

Post your results


tps1layout.jpg


tps1componentview1.jpg


1698276438200.png
 
Last edited:
Looks something like the GT power supply schematic.  The schematic in the link below does not  match the pictures more like right parts wrong order.
http://www.twin-x.com/groupdiy/albums/userpics/GT_PS-1A.gif

I see some issues, as RuudNL posted the zener looks like it is running hot.  One of the 220uf 160VDC caps heat shrink looks like it is pulling back.  No led current limiting resistor (maybe in line with one of the LED legs in shrink tubing?)

I looked up the bux85 dc current gain of 30 to 50 at 100ma I don't see a hfe vs Ic curve at the link below so we don't know the hfe at the lower microphone current.  One would need to measure it at about 1ma as a first guess.
http://www.datasheetcatalog.org/datasheet/motorola/BUX85.pdf
http://www.datasheetcatalog.org/datasheet/microsemi/1n5333b-88b.pdf

Also 120VAC X square root of 2 = about 170VDC.  The first B+ cap after the diodes might be over voltage with no load.

The copper pad area at the zener leads could be increased in area to help a little as a heatsink. The zener could also be raised up off the board a little for more air flow around it.   You need just enough current under the microphone load to have the zener work AND enough current for the base drive.
 
I don't doubt it works fine. But maybe it could be finer?

Using nominal 160V raw supply, I get 1.77 Watts in the Zener. The headline says 5W, but Fig 1 shows 1.4W on PCB board (5W needs an infinite heatsink on the leads).

As Gus says we expect 168V raw if everything were ideal. With small iron it will be low, but modern electric distribution often runs high.....

I feel this is WAY too close to early death. 'Specially since it would be easy to fix.

> I don't see a hfe vs Ic curve at the link below so we don't know the hfe at the lower microphone current.  One would need to measure it at about 1ma

Actually hFE could be as low as "2". What is the mike current? Maybe 1mA, and for the stated tubes probably not over 4mA (also we don't want more heat in the shell). What is the Zener resistor current? (160V-120V)/2700 = 14mA, split between Zener and Base. The 120V Zener gets a bit sloppy at 1mA (Zzk=1150) but we can go to 2mA, leaving 12mA to the Base, we only need 4mA to the mike.... it's not clear what this transistor adds to the party.

If you KNOW your mike current (and your wall does not vary 108-129V), a C-R-C-R-C-R-C dropper/filter is hard to beat.

If current (or wall-volt) may vary, it is efficient to run the Zener at a low current and use a transistor to buffer it. This also reduces waste heat on low-current or disconnected loads.

I don't see any real short protection. (Which is odd because there is a 9th 1N4007 which might be part of a current limiter, but it won't clamp until >1000V/28.9r or 36 AMPEREs!) The short current appears to be limited by an equivalent resistance Rz/hFE or 2700/50= 54 ohms, which at 160V is about 3 AMPs.

Layout: I like pre-made FWB modules whenever possible. Half the joints and fewer ways to put something backward (who does that?). I dislike the 12VAC and 120VAC wires crossing near the heatsinks, I picture it as untidy and cluttered where we may have to go in for repairs. PCB allows us to put the wire-maze on the board and keep the PT leads together on one edge for neat out-of-the-way routing.

The ground paths seem random, but at least the nasty path from rectifier to first cap is fairly local, not crossing through the clean output zone.
 
I drew the schematic out from the pictures
Raw DC to 1st 220uf and to one side of 1st 2.7K and 1st Transistor C
first 220uf cap - to ground
other side of the same 2.7K to the cathode of the zener and + of 2nd 220uf cap that node to the Q1 B
anode of zener and 2nd 220uf - to to ground
Q1 E to +1st 47uf, - of cap to ground

Voltage regulation and cap multiplier(cap value x hfe) circuit but like PRR posted not much gained if hfe is low

next stage
last node to C of Q2 and one side of 2nd 2.7K
other side of 2.7K to B of Q2
B to +2nd 47uf other side of cap to ground
cathode of 1n4007 to base
anode of 1n4007 to one side of the 27ohm
other side of the 27 ohm to Q2 E
emitter to + 3th 47uf cap - cap to ground
output of circuit from Q2 E

Cap multiplier again?

Simple thing to do is make the zener two or more zeners in series to spread out the heat.

fil supply part needs some help as well.

I also think a bleed resistor across the 47uf cap from Q2 base to ground might be a good idea.
 
Was thinking the BUX85's had some current regulation duty in combination with the 1N5380B Zener for voltage.

As for the filament side of the power supply , this is classic example linear regulation (7812).

Intention here was to give this away free as a sort of educational build. It is a working example and cheap to build for anyone wanting to DIY this at very low cost point. As a point of intererest, I found this power supply to be a different design in the HT 120vdc section compared to the typical R-C network & Zener , where you usually don't find power transistors in the circuit like the BUX85's. It certainly does attract attention to its design. As far as efficiency, I do not have any technical data or schematic, but its seems likely that the BUX's have some interaction here. The use of such devices opens new ideas of thought when operating the HT power supply section of a tube microphone.

nice to see all of the inquiries and responses, the technical suggestions are great points to consider
and implement to improve the existing circuit. Hope someone finds this power supply useful as is or the basis
for a improved design.



 
There is nothing new with that design.  Look in older books for series pass regulators etc.
IMO this is not a good design for a beginner DIYer
The heater supply section is lacking IMO.
The B+ part needs some work as seen by the discolored PCB under the zener.
One should look into a darlington or other devices for the pass elements.


 
I was thinking a LM317HVT linear regulator instead of 78xx for the filament since its easy to adjust for range different tubes.

Also , already mentioned a pair of 60V Zeners in series with 5W rating to share the heat.

Lastly
Here is a similar schematic with  pass transistor used after the second series power transistor. Not sure if this fixes some of the weak areas you indicated in the HT section. Would be ideal to have a adjustable HT
section here to manipulate plate voltages. How or what can Darlingtons contribute here ?

schem1.jpg
 

Latest posts

Back
Top