Building an energizer almost from scratch - advice on clamping zeners

[bobkatz]

Background: Just for fun, I'm going to build an energizer using a small Class-D power amplifier and a transformer. The inspiration was that I'm basically junking my recently-purchased and rejected Topping EHA5 and O can use the chassis and connectors. The rest I plan to junk. I bought a high voltage DC-DC converter for the bias voltage and a pair of toroidal transformers meant for a tube amplifier. The transformers are so big I may have to mount them on top of the Topping Chassis! For the amp I'm hoping that the Texas instruments TPA3116d2 Class D chip will do a good job. If I don't like the soun of this chip, I'm only out $10 for a populated board from Amazon!!! 

OK, if anyone can help, please, I'd like some advice on clamping Zeners for driver protection. Cribbing from a Stax SRD-7 schematic I see ZL01 through ZL04 but it doesn't give the values or the orientation of the Zeners. If possible, it would be nice to pick a value that will protect the headphone but have no sonic effect on headroom or impact. Many thanks in advance for any advice! 

 

[kevin gilmore]

you definitely want to use the lundahl transformers, they are going to be better than anything else you can buy.

[simmconn]

@bobkatzIMHO you are not likely going to get better performance with the TPA3116d2 class-D amp. It will have high noise and distortion than the amp modules in EHA5, plus the ultrasonic switching noise that may cause further intermodulation junk down the chain. Most if not all the imperfections you see in the measurement are caused by the transformer, such as higher THD at lower frequencies, the resonance at ultrasonic frequency that manifest itself as poor square wave response, plus other issues like phase shift (group delay distortion) that is not reflected in your tests. The transformer is the weakest link here, within the capability of the amp modules.

Using a better transformer such as lundahl would help. I'm not sure if you can find one with turns ratio as high as 50. If not, you might want to switch to an amp with higher output voltage, if you need to maintain the 700Vrms output level.

[bobkatz]

14 hours ago, kevin gilmore said:

you definitely want to use the lundahl transformers, they are going to be better than anything else you can buy.

I know. I have experience with Lundahl in another application. But I do have a source for another extraordinary transformer that I can afford that i hope will work very well. We shall see. 

[bobkatz]

9 hours ago, simmconn said:

@bobkatzIMHO you are not likely going to get better performance with the TPA3116d2 class-D amp. It will have high noise and distortion than the amp modules in EHA5, plus the ultrasonic switching noise that may cause further intermodulation junk down the chain. Most if not all the imperfections you see in the measurement are caused by the transformer, such as higher THD at lower frequencies, the resonance at ultrasonic frequency that manifest itself as poor square wave response, plus other issues like phase shift (group delay distortion) that is not reflected in your tests. The transformer is the weakest link here, within the capability of the amp modules.

Using a better transformer such as lundahl would help. I'm not sure if you can find one with turns ratio as high as 50. If not, you might want to switch to an amp with higher output voltage, if you need to maintain the 700Vrms output level.

Thanks. That's what I was afraid of. Keep in mind that the intermodulation distortion I saw in the Topping amp concerns me. I think that's the amp module, not the transformer as I don't think IM is a typical component of a transformer. If the TPA amp sounds and performs like crap I'm only out $10 each and I'll seek another solution. I'm thinking an alternate amp solution that I can handle and implement conceptually is a +-24 volt power supply with a pair of 990-style opamps in push pull. A 990 can deliver 13.8 volts RMS into 75 ohms so it's a pretty powerful discrete opamp. I found a 33:1 transformer so basically if I want to get to max of 600+ Volts RMS/sine wave I need nominally up to, say 19 volts RMS in. That's 19.8 volts in push pull. And as you all know, push-pull helps cancel even order harmonics, which could help the purity of tone. 

Again, if I don't like the 990 approach, I'll try something different until I'm happy with the sound 🙂

 

Back to the clamping zeners. Does anyone have a value recommendation and a simple orientation schematic? Thanks!

[bobkatz]

OK, I think I found the zeners, it would be wonderful if you would check my idea:

1) Protect Bias supply from over voltage. A single Vishay 1N5061, 600 volt avalanche diode. The supply I bought on Ebay is a DC-DC converter, Bertan, adjustable high voltage power supply, up to 1 kv at 4 ma. Takes in +24 volt. So it would seem a good idea to put the zener across its output. 

2) Protect the headphones. How about a pair of 800 volt Vishay 1N5062. What do you think?

[simmconn]

An ideal shunt-type protection device for the estate phones should have:

1) Low (<10pf) and constant capacitance vs voltage and vs frequency - that pretty much crossed out all the silicon junction devices.

2) An I/V curve with a sharp corner, near-zero leakage within the operating voltage and well-controlled clamping voltage - that rules out most MOV devices

3) Fast acting. That excludes the gas-discharge type.

What do we have left?

Why use a shunt-type protector, just to give the amp and the transformer a jolt like a dead short when they are outputting close to their max voltage? The Stax transformer box adds a PTC in series with the transformer primary to mitigate this issue - and you know how people hated it.

IMHO a better protection is to limit the output voltage of the amp. If the power rail voltage is limited, the amp's output is limited and the transformer's output is limited as well.

Edited July 30, 2023 by simmconn
Fix typos

[bobkatz]

14 minutes ago, simmconn said:

An ideal shunt-type protection device for the estate phones should have:

1) Low (<10pf) and constant capacitance vs voltage and vs frequency - that pretty much crossed out all the silicon junction devices.

2) An I/V curve with a sharp corner, near-zero leakage within the operating voltage and well-controlled clamping voltage - that rules out most MOV devices

3) Fast acting. That excludes the gas-discharge type.

What do we have left?

Why use a shunt-type protector, just to give the amp and the transformer a jolt like a dead short when they are outputting close to their max voltage? The Stax transformer box adds a PTC in series with the transformer primary to mitigate this issue - and you know how people hated it.

IMHO a better protection is to limit the output voltage of the amp. If the power rail voltage is limited, the amp's output is limited and the transformer's output is limited as well.

Limiting the amp's power? That's a double edged sword! My instincts say that limiting the power rail just means the amp is going to clip earlier and to my ears, the closer you let an amp get to clipping, the harsher it can sound. So I'd like to keep the amp well below its clipping point. Now it seems to me if you put a current limiting resistor, like a 5k, or two 2.5K on each audio leg, in series with the leads to the phones, then the zener after that won't be shorting the amp and transformer when it clamps with an overvoltage. 

But I can do a sort of protection by carefully adjusting the analog gain structure of my chain post DAC so that maximum analog level (top of the analog volume control) at 0 dBFS (maximum digital peak level) will produce no higher than 700 Volts RMS at the secondary. First of all, that would be very loud and my analog gain would normally be well below that. So I would view the zener approach as just an emergency clamper, a protection from unforseen circumstances. How does that sound?
 

[bobkatz]

Is using like a 2.5 kohm in series with each leg to the phones a bad idea? Is it inviting high frequency losses? Instead of feeding the transformer directly to the connector pins?

[simmconn]

2 hours ago, bobkatz said:

Limiting the amp's power? That's a double edged sword! My instincts say that limiting the power rail just means the amp is going to clip earlier and to my ears, the closer you let an amp get to clipping, the harsher it can sound. So I'd like to keep the amp well below its clipping point. Now it seems to me if you put a current limiting resistor, like a 5k, or two 2.5K on each audio leg, in series with the leads to the phones, then the zener after that won't be shorting the amp and transformer when it clamps with an overvoltage. 

But I can do a sort of protection by carefully adjusting the analog gain structure of my chain post DAC so that maximum analog level (top of the analog volume control) at 0 dBFS (maximum digital peak level) will produce no higher than 700 Volts RMS at the secondary. First of all, that would be very loud and my analog gain would normally be well below that. So I would view the zener approach as just an emergency clamper, a protection from unforseen circumstances. How does that sound?
 

That's all good statements in theory. My suggestion would be to plug-in the real-world numbers, do the math and find out what kind of voltage you really need/can afford, and go from there.

For the protecting scheme, too. You now have a high voltage DC supply which can conveniently verify the protecting threshold of the candidate device. You could also connect the protection device/series resistor(s) to your phones and listen to them at comfortable levels. If you can't hear a difference, or even prefer the sound with the protection device present, viola! you've got a solution.

FWIW, I found an MOV which looks like a 2W+ metal film resistor in my old Stax Lamba (normal bias). I guess at the time the non-linearity it brings was the lesser evil compared to a burnt driver.

[dsavitsk]

On 7/29/2023 at 9:34 PM, simmconn said:

@bobkatzI

Using a better transformer such as lundahl would help. I'm not sure if you can find one with turns ratio as high as 50.

LL9202 can be wired 100:1

http://www.lundahl.se/wp-content/uploads/datasheets/1620_3_7_9202.pdf

[bobkatz]

I pulled the DC-DC converter from the Topping board to measure it and see if it's worth keeping for some purpose. It's an unmarked black potted module. Output voltage is somewhat proportional to input. It normally receives +15 VDC from the motherboard. I fed 15 VDC into the converter and got 591 volts out, no load. Unfortunately, the higher the input voltage, the worse the AC voltage (output noise or ripple) as measured on my Fluke 289. At 15 volts in, 591 volts out, the fluke shows a horrid 2 VAC. Without scoping it I can't tell you what kind of noise that is, but for me that's cause for concern. 

 

 

Bob

Edited July 31, 2023 by bobkatz

[bobkatz]

5 hours ago, simmconn said:

That's all good statements in theory.

I can put a square wave into the transformer and see what level trips the Zener and how quickly. And the effect on the system. Given that it's the high impedance winding, even if i don't put in a resistor, it would seem to me a kind of current limiter.... just a thought, not my field of expertise. 

[bobkatz]

This topic has stretched from just zeners to another OT transformer energizer build thread :-). I apologize if I'm going over old ground and if any of you head-case old timers can point me to threads discussing building a passive energizer I'd appreciate it. 

Next question: Transformer rated watts. Isn't a 15 watt transformer under rated in terms of its insulation voltage breakdown? My calculations are that with a 5000 ohm to 4 ohm ratio, that's nominally 35:1 voltage ratio. 15 watts at 4 ohms comes to 7.7 volts RMS with sine wave. But 35 x 7.7 volts ----> only 269 volts on the high impedance secondary. If we desire, up to, say, 600 volts RMS on the secondary, aren't we overdriving the transformer by at least 2X its original voltage rating?  I'm not worried about watts because the headphone load is very high impedance. But again, what about the 

My next question comes again back to my choice of amplifier. What is the reflected impedance back to the 4 ohm tap? In theory it's VERY VERY high because the headphone load is extremely high impedance. Even though the DC resistance of the 4 ohm tap is maybe 0.1 or 0.2 ohms. At least I need to be capacitively coupled. These opamps I'm considering can drive down to 75 ohm loads. Am I going to run out of drive capability with a discrete opamp driver?

[bobkatz]

I wish I had just titled my main thread "Building an Energizer from Scratch" as I have more questions than just about the zeners. Thanks for your understanding!

So, now I've decided to use a linear Class A amplifier board to drive the transformers. I think that's a better idea than push pull opamps or Class D noisy stuff. 

So, a question about heat sinking:

Taking a chance, I ordered two of these "JLH 1969 Cass A Amplifier" boards from Ebay. Just search for the description on Ebay "JLH 1969 class A amplifier stereo high quality PCB components assembled !" as there are tons of Chinese sellers. Hopefully I picked a good seller. Anyway, We'll soon see how they perform. I can measure and I have load resistors. About the heat sink required for these boards. For Stat headphone purposes, do you think just screwing the aluminum plate of this amplifier to the sides of the Topping amplifier chassis will provide sufficient heat sinking? 

Edited July 31, 2023 by bobkatz

[bobkatz]

I know very little about power amplifier design. I only know how to bias and other things by instructions of the designer. I picked this one based on instinct and that it can be adjusted to Class A. Is it such an old fashioned design that it's not going to sound good? Schematic is attached. I might change some of the caps to film in due time. 

[bobkatz]

Apparently this is some kind of legendary circuit, based on John Linsley Hood's design going back to 1969, as described in a Wireless World article: 

https://keith-snook.info/wireless-world-articles/Wireless-World-1969/Simple Class A Amplifier.pdf

 

Well, at least I picked an amp with a good pedigree. I'm sure there are bipolar symmetrical designs of this amp floating around. To my mind the weak links could be the electrolytic coupling caps in this single ended supply design. We shall see. 

[bobkatz]

BOY DID I MESS UP!

Good thing I discovered before I started building: I didn’t consider the output voltage swing of the power amp based on its supply voltage  

The JLH amp Can take up to 33 vdc supply at 3 amp. Optimistically lets Say 2 volts loss in the output swing

OK, output swing 31 volt p-p

15.5 v peak 

10.95 v rms sine wave

33x transformer ——->

361 volts out. Shit. 
 

Maybe back to plan A:

two opamps in push pull. I have to see if they can deliver up to, say 2 watts  conservatively based on Dr  Gilmore’s 1 watt 106 dB Stax estimate 

 

[bobkatz]

And what nominal impedance does the primary of the transformer present to the opamps or the power amp? Since the opamps can drive 600 ohms or maybe less, is it safe to assume the load will be higher than 600?  In push pull does each opamp see twice the nominal load on the transformer?

[simmconn]

Here is the impedance plot (phase and magnitude) of the transformer in an SRD-7SB with the SB circuit disconnected. Three sweeps were run with 1V input, output open; 1V and 2V input, driving a Stax SR-404 with its bias supplied externally. Take an educated guess, which curve is from which test condition?

[kevin gilmore]

worse than expected.

would like to see the lundhall transformer.

 

[bobkatz]

19 minutes ago, kevin gilmore said:

worse than expected.

would like to see the lundhall transformer.

 

I'm not prepared to take an educated guess. But I am prepared to tell you that a very smart consultant I know very well suggested getting negative feedback from the secondary of the transformer to linearize its performance. And that's the implementation I am going to try. I have two transformers, one is an excellent toroid that I found and the other is an unexpected transformer that is very available. If it works, then it will revolutionize this portion of the Stat market. We shall see 🙂

[kevin gilmore]

the phase is going to mess with the feedback in a serious way. going to be very tricky to get right.

and then there is the problem of disconnecting the headphones with audio playing.  might turn into a high power oscillator.lots of magic smoke.

[bobkatz]

Thanks for that advice. Crossing fingers

[simmconn]

2 hours ago, bobkatz said:

I'm not prepared to take an educated guess. But I am prepared to tell you that a very smart consultant I know very well suggested getting negative feedback from the secondary of the transformer to linearize its performance. And that's the implementation I am going to try. I have two transformers, one is an excellent toroid that I found and the other is an unexpected transformer that is very available. If it works, then it will revolutionize this portion of the Stat market. We shall see 🙂

I wish you doubled success with those. Oops, squared!😆