how do I use a 50v bias tap for a 48vDC mic phantom power?

[mike_mccue]

Hello,

I prepared a schematic for a power supply section with the intention of using a 50vAC "bias" tap to supply 48vDC as phantom power to a microphone.

I've only just learned that the transformer I selected is wired differently than I anticipated and as I had drawn in my schematic. I need to ask for help in understanding how to hook this up.

Here's my schematic (sorry for the size... I wanted to keep the lettering legible) that I know needs to be corrected:

EDIT: this schematic has been replaced with a more correct version as shown below:

So I am not sure what to do. Does the supply to the 5y3 need a center tap to ground? Can I just use the center tap and the 50vAc as my two connections to the solid state bridge rectifier sort of like I have shown? Or do I need to keep the center tap on the 275/275 connected to ground and float the 50vAC tap some other way?

Thanks for any help you can offer.

best regards,
mike

[6267]

In the original applications, the tap would have been attached to a halfwave (one diode) rectifier in say a bias circuit. It would never have had to provide appreciable current. You could emulate that, although it will take more filtering, the zeners & choke should regulate OK. Make sure to account for fault conditions which might attempt to draw mucho current from the 50v winding, since it'll probably die quickly under heavy load and take out the B+ (and thus transformer) with it. Bad cables, etc...

[mike_mccue]

Thanks,
If hooked up properly the draw at 48vDC will be apprx 8ma total for a pair of modern "transformerless" condenser mics.

It can be as low as 2ma (older transformer equipped mics) and as high as 20ma (newer DPA or Earthworks mics). Any needs beyond that will be, as you describe, caused by some sort of fault.

So is it OK to float the 275 leads to the 5Y3GT? Can I use the center tap for the other half of the bridge rectifier?

thanks,
mike

[mike_mccue]

I found this at a wiki about rectifiers:

"in a circuit with a non-center tapped transformer, four diodes are required instead of the one needed for half-wave rectification.[/B] (See semiconductors, diode). Four rectifiers arranged this way are called a diode bridge or bridge rectifier:

For single-phase AC, if the transformer is center-tapped, then two diodes back-to-back (i.e. anodes-to-anode or cathode-to-cathode) can form a full-wave rectifier.

A very common vacuum tube rectifier configuration contained one cathode and twin anodes inside a single envelope; in this way, the two diodes required only one vacuum tube. The 5U4 and 5Y3 were popular examples of this configuration."


I'm inferring from this that the 5Y3GT MUST be hooked up to a center tapped secondary... as I originally planned and drew up.

So, Is there some way to float the 50vAC so as to hook up a bridge rectifier for the phantom power?

thanks very much,
mike

[Enzo]

First, the 50v winding. It is going to rectify to 70VDC, and I wouldn;t want to just slap four zeners across that and expect 48v with nothing more than a choke between them and the 70v. I would suggest adding a dropping resistor and additional filter stage.

The replacement transformer layout is common. The HV center tap is grounded, and the tap is 50v away from that. What the rest of the HV winding does is irrelevant. The 5Y3 must be wired in the manner shown in your schematic, yes with center tap. I would just wire the 50v tap exactly as you would for a bias supply, except reversing the polarity to positive. There is very low current draw for the phantom mics as mentioned above by Mike. SO a half wave rectifier and plenty of good filtration and voila, you have phantom.

[octal]

What's the deal with that 500 uF first filter cap for the main B+? Seems like a recipe for blowing rectifier tubes. I imagine that input choke will help a bit but....

What do other folks think about it?

[defaced]

And the 68v caps on the bias tap. Those should be at least 100v.

[Enzo]

The first element in that B+ is a choke, not the cap. That choke is what the rectifier tube will see. But you are right, I would much rather see maybe 100uf there. Especially since nothing uses that node anyway.

Amen on the bias caps.

[mike_mccue]

Thank you everyone,

I'll edit the drawing to use the half wave rectifier circuit.

I'll upgrade the 100uf 68vDC Caps to 100vDC.

Is the concern about the 500uf Capacitor that the inrush mat be too much at power up?

I'm used to the smaller uf sizes that we see in guitar amps. The 500uf spec is inspired by an enthusiasm for bass and stereo amps... and seemed appropriate for a microphone preamp.
Is it being suggested that I use a single 100uf cap at the front and the 500's beyond or are you suggesting I ditch the 500's all together?

Last night I was learning about rectifiers choices at the Hammond site. I didn't fully realize that each design resulted in drastically different levels of filtered DC.

I have to look at the B+ and see if the choke be ok... it may drop too much voltage.... or at least that's what I think I learned last night.

thanks again for all the insights and suggestions,
mike

[defaced]

Yes, a choke input will drop your B+ significantly. Steve Conner has also mentioned several times on here that in order to properly do a choke input amp you need a "swinging choke". I don't know what that is, but he has said they are not made anymore. You will likely want to research this, I have been wrong in remembering stuff in the past.

This thread might be of some help: Overfiltering?

[mike_mccue]

Here's the latest schematic with revisions:

C04 and C05 are now specified as 100vDC and I've begun working with the half wave design:



After a quick read through Hammond's rectifier page I was left with the impression that the choke mainly reduced the peak voltage and that the average voltage was the same. At this point I feel a bit confused... should I place the choke before the cap? After? I have to go learn the implications.

I'm somewhat familiar with the idea of over filtering because my previous tube amp experience is with small champ style amps where the distortion and sag is appreciated and over filtering is to be avoided.

I'll go read that thread to learn more.

Please keep the comments and observations coming. :-)

thanks and best regards,
mike

[Steve Conner]

Yes, as per what Enzo said, you need some resistance in the circuit to let your zeners do their work without burning up. (With just the choke, I'd expect the zeners would burn up, fail short, and burn out your transformer's bias tap.)

You might as well replace the choke with a resistor, and if this causes too much ripple, put some resistance in series with the diode too.

I've also used LM317/337 regulators hanging off bias taps:
http://scopeboy.com/bias.html
http://scopeboy.com/bias.gif

(Notice the fuse to save the bias tap if anything bad happens.)

[mike_mccue]

Thanks again everyone,

I went and read the over filtering link. That was a great read. I had no idea that there was a well known concern about Tube Rectifiers and over filtering. I'm glad to have learned of it.

Please let me explain the nature of this project. I'm hoping to construct a dual mono tube microphone preamp. My design goal is to keep it very simple and while I want to use good engineering I also want a very traditional design and the resultant sound... at least in an arbitrary sense that I want to "date" the design. For example I have been insistent on the tube rectifier because of my interest in sag.

I see that it's obvious that the 500uf caps came straight out of hi-fi land :-). My influence regarding those was a bass amp and hi-fi aficionado whom has been a good mentor. He recommended a solid state rectifier. When I insisted on a tube rectifier the details about the implications of over filtering slipped through the cracks.

Thanks very much for the help with that.

In response I've edited the schematic to use (to the best of my understanding.) a Pi filter on the hi voltage secondary. I've replaced C1 500uF, 385vDC with C1b 100uF, 500vDC and added C1a 100uF, 500vDC:



I just spec'd the 100uf 50vDC rating by familiarity rather than math. Did I do ok?

I don't not specifically have a RL as seen in the Pi filter explanations I just read... does it matter? Is the circuit itself the RL load?

Is this an acceptable solution or compromise for my goals?

@Steve

I'm off to read your stuff... but I was a bit confused by your response. Not in general but in specifics. I assume you are speaking of protecting the IN4007? But maybe you meant the IN4742s? I'm outside of my experience designing the power supply for the 48vDC.

There is one concession to my interest in a traditional circuit. I want a top of the line super stable and smooth 48vDC phantom power supply.

I have made the previous edits to change to a half wave rectifier... I'm concerned that the ripple just went up.

In all honesty the choke was placed on the 50vDC sort of gratuitously... I know it is overkill but I already bought it :-).

I don't know much about modern voltage regulators and would enjoy learning anything I can about laying out a nice 48vdc circuit.

If the choke has to go I'm open to it. I just need to learn what's best.

FWIW I have incorporated the precautions noted in the Jensen transformer white papers regarding connecting the 48vDC to the mic input portion of the schematic.

thanks again and best regards,
mike

[Steve Conner]

Yes, I meant protecting the 1N4742s. If this confuses you, you should study how zener diodes work.

If you're looking for something to do with that second choke you bought, why not put it in series with the first one in the main B+ line.

Or leave it out altogether, you're in the running for "Heaviest And Most Overkill Mic Pre Ever" as it is. A "top of the line" phantom supply would be the LM317, IMO.

[mike_mccue]

Thanks for the advice. I appreciate the candor and the benefit of your insights.

I am primarily a big fisted point to point wiring solder DIY hobby person... so I am inexperienced working with smaller components. I guess I design around that. :-)

This project is a long term dream of mine... I want to hear the sound of a super simple tube mic preamp amp with a monster power supply... you've identified the symptoms accurately :-)

I went off and learned a bit about the LM317 series of voltage regulators.

I only found one listing for a version that would supply 48vDC. Here's what I came up with:



Steve, your page was very helpful. I included the fuse before the IN4007 from your design. As I said I was a bit confused... did you also specifically recommend the small resistor between the IN4007 and the LM317 in a previous post in this thread?

I think I made the proper calcs for the R1(a.k.a. R22) and R2(a.k.a. R23) resistors on the LM317... can someone check me on that? I became somewhat confused about the Iadj. I saw some calculations disregarded this as being a inconsequential factor... and treated it as near zero. Is there a proper value to assign Iadj while solving for R1/R2?


I am also curious about the hi voltage power supply... I think I have predicted the correct voltage values on the secondary now that I have placed the 100uf 500vDC cap at the beginning of the filter stage. Does it appear that I'm making that estimation correctly?

Thanks very much and best regards,
mike

[mike_mccue]

bump... in hopes someone can offer some comment. :-)

[tubeswell]

I am also curious about the hi voltage power supply... I think I have predicted the correct voltage values on the secondary now that I have placed the 100uf 500vDC cap at the beginning of the filter stage. Does it appear that I'm making that estimation correctly?

Hi Mike

I've never seen a 12Y3GT datasheet, but if you can find one, it should tell you what the current output of that tube is and the voltage drop as well as the maximum recommended reservoir capacitance (intuitively the 100uF seems to be too high - the 'Y3GT' nomenclature suggests to me it is likely to be in the same ballpark as a 5Y3GT, for which 40uF is pushing it - but I could be way wrong about this). Aside from that, I'm sure you are aware that with the CLC filter you've changed it to, you don't need as-big-a-choke to get to your desired B+ (as you would with a choke- input filter), and your B+ will also be higher. Using Merlin b's rule of thumb for CLC filters you could even use 20uF reservoir and 20uF output cap with a 5H choke and get excellent filtering results. JM2CW

[Steve Conner]

OK, here are some comments:

I don't believe there is any such tube as a 12Y3GT. And if there was, it would have a 12V filament, so it wouldn't work off a 5V heater winding. You probably meant a 5Y3 of some sort. For a tube mic preamp, the 5Y3is probably overkill, the 6X5, EZ80 or whatever is probably more suitable. Unless you're going true vintage style, with a push-pull line output stage powered by 6V6s, that puts out enough voltage into 600 ohms to double as a cattle prod.

The 70V DC is still over the rating even for a LM317HV. This means that if you short the phantom power, the LM317 will blow. It will most likely fail short and put 70V onto your mic. You need a zener diode or dropper resistor in series with the input of the LM317, to drop some voltage. I showed how to do this in the schematics I posted earlier.

Let's check your calculations for the LM317 resistor values. First calculate the adjust pin current with Ohm's law: 1.25v/220 ohms = 5.68mA.

Now multiply this by the other resistor: 0.00568 * 8228 = 46.75V.

Finally add 1.25 to get the actual output voltage: 48V. Yay!

[mike_mccue]

Thanks!!!

I fixed the 12YGT typo to read 5YGT. Thanks for catching that.

I'll consider the comments and be back later... I just wanted to settle the issue with the typo.

best regards,
mike

[mike_mccue]

I added the dropper resistor to the schematic:



Now I'm off to re think the hi voltage power supply.

Thanks again everyone,
mike

[tubeswell]

Well with a 5Y3GT off 270VAC-0-270VAC, you'll only get something like 300VDC. I would only go to 40uF max with a reservoir cap for your CLC filter with this tube. But if you use a GZ34, you'd get around 350VDC and you could easily go to 60uF for the reservoir (2 & 1/2 cents worth this time)

[mike_mccue]

Hi,

It took me a while to get back to this. Thanks for the reply.

I'll spec C01a at 40uF.

How about the other filter caps? I realize 500uF caps are excessive but had planned for them to entertain the notions of a previous mentor who had an interest in stereo and bass amps. I do have the 500uf caps here and paid for, but am willing to downscale if that's for the best.

thanks again,
mike

[mike_mccue]

Hi,
I'm just bumping in hopes that anyone with better sense than I can offer, repeat, or make note of any un acknowledged opinions before I start drilling into a chassis and building a layout board.

:-)

Thanks again for all the help.

best regards,
mike

[defaced]

I'd probably go overboard and make the 100 ohm resistor 2 watts. Right now you have a 100 ma fuse, assuming the resistor will see the full 100 ma in a fault condition, it'll dissipate 1 watt. Since you generally want a margin of error on power ratings, and the next common size is 2 watts, I'd use 2 watts.

As for the 500 uf caps, I'd at least try them. They'll probably be pretty big physically, so you may want to get something smaller that will work just as well.

[mike_mccue]

I had come up with the same reasoning... I'll spec a 2 watter for now.

thanks,
mike