Does anyone care to ask an amplifier designer a technical question? My door is open.

Hi ramtubes,

Thanks for answering my previous question.  I appreciate it.

I do have another question please...

I am building a symmetric plus/minus 70 volts dc (2 x 50 volt ac secondaries) power supply for an amplifier I am building.  Mechanical issues are giving me more trouble than electronics such as drilling holes accurately etc.  To this end, I want to build this so I do not have to change the smoothing capacitors for a LOOOONG while so I dont have to re-drill holes etc if and when they break.  I am concentrating on those specific capacitors since they are large electrolytics and hence are most likely the least reliable of the components.

In your experience & opinion, which make (and series/types) of such large electrolytics are the MOST reliable in the long term ?  I am considering about 40 to 80 milli farads at 100 volts per each side of the supply, depending on the price & physical dimensions.  I predict that these capacitors will be in a vicinity with a temperature of about 60 - 65 C most of the time.

Thanks

@bdp24 

Roger, I think you provided this information somewhere, but I can't remember where, and can't find it. So let me ask you: what is a good output impedance number to shoot for in a power amp? How low does it need to be to prevent frequency response peaks and dips due to the speaker impedance/amplifier output impedance interaction? I know the figure will be different for a speaker with a wild impedance curve (the original Quad ESL) than for one with a fairly even curve.

Great Question...  Damping of 10 which is an output impedance of 1/10 the load is a nice number and about all one can achieve in a tube amp and maintain staility. Quad achieved 20 on the model II for the ESL57 because Walker wanted 20, he said so in the speaker spec. The II is a fairly low power ampifier, yet a great one.

Futterman had a damping factor approaching 100 perhaps more. It has lots of well applied feedback which it can tollerate due to no transformer and its excellent internal bandwith (open loop bandwidth). I think it is clear that he was interested in low output impedance.

The open loop response of a tube amp is almost entirely dependent on the output transformer. When you see those peaks after 20Khz in JAs measurements that is all from the transformer actually reversing phase and making the feedback positive. Great indication of an amp that will oscillate into certain loads. Ringing on the square wave shows the same thing. 

Since you brought up the Quad 57, Walker or Baxandall did a very smart thing in the transformer to keep the impedance from going to infinity in the bass. They put 3 shorted turns of #16 wire, very loosely coupled inside the transformer. I only know because I melted the wax of one. *(by accident)* This clever idea helps tame the bass if an amplfier had very high output impedance. Even protecting the amp from overvoltage at low F so as not to arc the amp's transformer. This may take some further explanation. I had to sit down when i saw it.

Of course SS amps can have much lower output impedance, approaching that of a  heavy speaker cable or even lower than that cable. JA measures with a few feet if realistic cable.

As I write this, every time I write "damping factor" I have to back and change it to either output impedance or regulation.  I think I left it twice in the hope of shifting the tide a little bit.

The Futterman is an amazing amplifier I hope to cover in depth in the future. Waiting for questions to arise.

@cakyo
..

I am building a symmetric plus/minus 70 volts dc (2 x 50 volt ac secondaries) power supply for an amplifier I am building. Mechanical issues are giving me more trouble than electronics such as drilling holes accurately etc. To this end, I want to build this so I do not have to change the smoothing capacitors for a LOOOONG while so I dont have to re-drill holes etc if and when they break. I am concentrating on those specific capacitors since they are large electrolytics and hence are most likely the least reliable of the components.

In your experience & opinion, which make (and series/types) of such large electrolytics are the MOST reliable in the long term ? I am considering about 40 to 80 milli farads at 100 volts per each side of the supply, depending on the price & physical dimensions. I predict that these capacitors will be in a vicinity with a temperature of about 60 - 65 C most of the time.

Some sellers are asking big money, just wait for an auction like this.

eBay item number:192627561898

Just an example. I had some 49,000 at 75 V, look for something like that. Do you have your power transformer yet? Hope its at least 6-10 amps. How do you intend to use this power?

10,000uF 100V Sangamo Electrolytic Capacitor Computer Grade Crown Power Amp BGW. These went for $8 each. No problem with age or use, they last forever. Get a few extra at this price and never worrry. Besides the diameters are pretty standard in only a few sizes so there is a lot of interchangeability.

Mallory, Sprague and other US makers made these. Its sad how many of these I had  to thrown away. Tried to sell them on eBay, no luck. Which is good for you as there are people who will let them go for nothing. I see one for sale now fot $380.. he's a newbie. :)

@wolf_garcia

On another note, neither of my current fave tube guitar amps has a "standby" switch as they’re relatively low powered (18 and 15 watts respectively), so the question is: When can I get my beer? Another question I’ve asked elsewhere is how do Premium Fuses work? What do they do and why? Thanks in advance for the answer (or answers).

The standy switch on a small amp doesnt save enough to pay for it. When you have 4 x 6L6s and a 5U4 then it makes sense.

Premium fuses don’t work. They don't protect wery well. They have caused a great deal of damage in DC circuits like the tube fuse in my RM-9. Hi Fi TuningFuses demonstrated their complete lack of knowledge by putting the fuse wire in a teflon sleve to "reduce microphonics"... Fuse microphonics, how cleverly stupid. The teflon containes the plasma which is the opposite of what a sand filled fuse does. The writers of the literature come from the bottom of the barrel with no knowledge of electronics. Just ask them sometime.. I have.

Read my first sentence in the OP. I had to shut down a fuse thread because we were not getting anywhere. I realized there is no logical argument that will sway them an inch. The Fuse evidently makes them feel a certain way. 

My advice stick with Buss and LittleFuse. They know what they are doing and won’t rob you like those Premium fuse peopls. They are all theives preying on the unwashed audiophile. Fuses bring out trolls perhaps thats how they work.

I am truly sorry anyone ever thought of Premium fuses.

@maplegrovemusic 

Roger - I own a pair of king sound - King v1 and a pair of King sound ks17 . This direct drive amp you spoke of , how does that work ? do I have to bypass the step up transformers , and do the speakers need to be modified to use that amp ?

The advantage of a DD amp on ESLs is to completely remove the enormous step up transformer and use it as a doorstop. 

As I have written here before, I have found cases where the step up is more than half the load.  Try to find the previous posts where I get more into it. 

If one plays most large ESLs on reasonable volume trumpet music they need 1500-2500 watts per channel. Roger Sanders knows this, I have measure this, there is no doubt about this. Beveridge, Acoustat and Sanders use full panel rather than segmeted panel like  QUAD. Thus a QUAD can be driven by 100 watts. So you see not all things are the same. 

What technical data can you provide me on  your speakers, Capacitance, impedance, polarizing voltage......