Design Me A Transformer, I Will Pay

people like me do when they need something done they can't do ? Pay someon e.

to source it so custom mad, oh well. You get what you pay for.

need a low DCR.

ts into one end and have -30 volts come out the other end, able to feed low impedance. I need it to do that efficiently at least from 20-20,000.

d so just 5% is a Cnote. If you're any kind of an audiophile I could just g ive you one of these amps. There really is nothing like them on the market.

il, and of course dot com. You can reply here, just how much do you want to do this ? I need wire size, number of turns and the core specifics. I know the basics about transformers but this is beyond me. Frequency response fl at and all that, I have no clue. An RF transformer I could tune with caps, a power transformer, actually I tried using one once and the performance wa s not good.

good results. The 4 is the center tap and the 0 and 16 are the ends. thing is if I go too high in power I am inducing a hell of alot of voltage into t he "primary" which isn't really anymore. But it can arc and break down insu lation. I would consider them but for that, I would need them in the higher wattage range. Without the "primary" it should save money and weight.

to the tone controls, at the highest turnover frequency the treble will ha ve +/- 20dB range, if the turnover is lowered by the control to boost more of the spectrum I want to decrease the range so people don't blow their twe eters. I might make it switchable - "treble protection". Other than that ov er half the math is done and I am all the way back to the tone board, it is designed but I want a differential drive to the output circuit.

There are many hundreds of custom design houses, and they do take orders fo r prototype small quantities like onesies. They have the manufacturing down to a triviality, usually working out the details in real time on their in house CAD in a few minutes of phone conversation. You just need to know wha t you want and be prepared to answer their questions about the application. Looking their quote spec sheet over beforehand should help. The product do esn't cost that much at all. Here is one as an example, I've never used the m and don't know anything abut them- they're just an example:

Try todd newava.com. I recently sent him a spec for a special toroidal power transformer and he came back with a nice proposal and decent pricing. He's done some good stuff for me before.

You don't need a transformer, you need power amplifiers capable of 60 Vp-p and 15A. That's 900 watts. Or maybe two 450-watt amps. There are class-D designs that can handle that. In the event you have slightly miscalculated, there are Chinese sellers of souped-up class-D chip eval designs that can handle those power levels, selling for quite low prices on eBay. I have purchased and received a selection, and they're not so bad. Transformers suck.

--
 Thanks, 
    - Win

** Hummmm - use a class D chip amp or two plus no doubt a high falutin' SMPS *instead* of a simple iron core transformer.

That IS a damn novel idea.

.... Phil

On a sunny day (6 Sep 2019 19:12:15 -0700) it happened Winfield Hill wrote in :

I presume you mean 60 Veff??

(60 / (2 * sqrt(2) ) ) * 15 = 318.198052 W

Wpeakpeak?? ;-)

(Have not read the OP)

----------------------------------

** Win is off with the fairies, again. 60Vp-p = 21.2V rms.

With a 4 ohm load that is 112 watts.

** Or even woken up ?

...... Phil

Dunno what the O.P. is doing, but I always like to think in terms of the max DC power. Certainly when using a class-D MOSFET stage, you're always considering the Vmax and Imax. Pmax Pavg Pwhatever, posh, who cares?

--
 Thanks, 
    - Win

TPA3255 looks great. It's good for 600w mono or 300 stereo, for under $10. I've just got to figure out an elegant way to heat sink it.

I can get a MeanWell 600 watt 48V power supply for something crazy like $80.

If I understand your configuration correctly, you have an autotransformer driven in a class B connection at the ends. The transformer will need to have the thinnest laminations you can get in order to handle the 20kHz. You will need to have enough laminations (core) to not saturate at 20Hz. But the designer should know that. You will need a wire size to handle the current (about 10ARMS) without generating excessive heat and/or without excessive resistance.

I suggest the following specifications for your transformer:

Autotransformer (bifilar winding)

20Hz - 20kHz Current - 10Arms Voltage rating is not a problem. Even the wire enamel can handle it. Suggested lamination source: Thomas & Skinner (No connection with them).

I designed transformers, but have no experience with audio.

By the way, this is my understanding of your topology:

Version 4 SHEET 1 880 680 WIRE 0 48 -208 48 WIRE 256 48 64 48 WIRE 368 48 256 48 WIRE 368 80 368 48 WIRE 256 128 160 128 WIRE -208 144 -208 128 WIRE 0 208 -208 208 WIRE 256 208 64 208 WIRE 368 208 368 160 WIRE 368 208 256 208 WIRE 160 272 160 128 WIRE -208 304 -208 288 FLAG 160 272 0 FLAG -208 144 0 FLAG -208 304 0 SYMBOL ind2 240 32 R0 SYMATTR InstName L1 SYMATTR Value 2 SYMATTR Type ind SYMBOL ind2 240 112 R0 SYMATTR InstName L2 SYMATTR Value 2 SYMATTR Type ind SYMBOL res 352 64 R0 SYMATTR InstName R1 SYMATTR Value 4 SYMBOL diode 64 32 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D1 SYMATTR Value RFUH30TS6D SYMBOL diode 64 192 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D2 SYMATTR Value RFUH60TS6D SYMBOL voltage -208 32 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value SINE(0 32 20) SYMBOL voltage -208 304 R180 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value SINE(0 32 20) TEXT 200 -16 Left 2 !K1 L1 L2 1 TEXT -434 328 Left 2 !.tran 0 1000m 0

If you check on eBay, you'll find lots of finished board choices, with heatsinks mounted, etc., starting at $36, free shipping. Some include toroid AC-line transformers, etc., and a case.

But the TPA3255 is less interesting to me, because it's single-supply, with ac-coupled input and BTL output. As opposed to split-supply types with DC coupling possible.

--
 Thanks, 
    - Win

I'll be driving a huge custom toroidal output transformer, so single-supply is fine. One concern is the output DC offset, which could freak out the transformer: The Devil's Staircase thing. I guess I'll add about 40,000 uF of electrolytics in the transformer primary, just to be safe.

We do have the TI eval board and have been abusing the chip. Haven't blown it up yet. I wouldn't want to base an aerospace test box on some no-name Chinese amp board. And I want to include some sensing and supervision stuff.

Some seem to be missing the point here.

e never used them and don't know anything abut them- >they're just an examp le:

Thanks, I'll check them out. But I have already been to a few companies and they don't seem to be interested.

dal power transformer and he came back with a nice >proposal and decent pri cing. He's done some good stuff for me >before.

I'll check them out.

You are not getting it, this is a brand new audiophile design. If I just wa nted power I could buy from B&O, those Icepower modules. If I wanted that i would not be here.

MPS *instead* of a simple iron core transformer.

B&O thought so and that is where the Icepower amps come from. Hundred watts no problem, no heat sinks. Well sure the copper on the board but the thing is smaller than an index card.-... that is about 10cm. by 15cm. The amps I have seen them in have little air flow and never a fan. They just don't ge t hot because they just don't waste energy.

I am building a solid state audiophile amp that most closely emulates the s ound of tubes, but much better quality. Class D is out of the picture. I kn ow what class D is and what it can do but it does not give me the sound I w ant. These things are going to go for two grand, and will be an all origina l design by me. I got the rest of it pretty much figured out but I lack the experise in transformers.

By bifilar do you mean the whole thing or insulated so I can wire it the ri ght way ? I need it to be equally efficient going both ways, there really i s no primary and secondary.

I have used output transformers for tube amps that had 0-4 and 16 taps on t he secondary, but that was inducing much voltage on the "primary" which was not a primary anymore. When i get to the power I want those will not work, the primary will short out by arcing.

I am BUILDING this.

Anyway, thanks for the responses. I got a couple more leads to try to get t hese.

The whole design is brand new. the outputs, drivers, predrivers, prepredriv ers all run on a resistive stack. The feedback is distributed, not global l ike most amps.

I could probably work this up on LTSpice as long as you can comprehend that L1 and L2 are one coil. It won't make any kind of simulation work right bu t it could illustrate the concept.

But do I want to reveal it ? This is totally new, I have seen more topologi es for audio amps than I care to remember. There is no off the shelf soluti on for what I want. There is no chip, no board, no nothing. In fact I might have trouble even finding suitable output transistor because they will be PNP. Darlingtons would be great, but i want the best of the best. Two grand each for these things, they have to be top notch.

Thanks for wasting your time on me. I appreciate it. I will check out the l inks.

Still, someone who knows what they're doing tell me how many turn of what g auge wire on a core of what size and composition, I would pay.

** Reading his posts might be a help ...

Not taking things so literally would be another.

** While ignoring all other information ? You are just adding to you original absurdity. ** Well clearly you don't.

Like I said already - way out of your depth and drowning.

Gurgle, gurgle gurgle....

...... Phil

--------------------------------

** Utter nonsense.

It's a wide band audio power tranny, not a SMPS.

** Right, full power at 20Hz happens all the time in audio ...

Strewth.

** If you did, you might know that a standard, tape wound toriodal core auto-tranny would meet the OP's spec with complete ease.

But you don't, so you didn't.

.... Phil

Yeah, Win is an amateur at electronics, compared to someone with your decades of experience fixing broken guitar pedals.

You should write a book.

It might weigh 100 pounds.

Phil, if you are so knowledgeable, then design him a transformer and take his money. Otherwise, shut the f*ck up.

Using .029 thick steel will not be good. What lamination thickness do you recommend?

I know, but what is Strewth? Is that a foreign language? Or maybe you are off you medications again.

Great! Suggest a core manufacturer and part number. Also suggest number of turns and wire size. That would prove you know what you're talking about.