gk71 Matrix amp

[Nick]

I don't see what the dcr had to do with it. The rectifier is not adding DC its passing a half sinewave

[IslandPink]

With utmost respect Mark you don’t say what is the dcr of you’re chokes and what rectifiers used or their dcr requirements, so how can youre anecdote help answer my question?

Well, 1. I didn't realise you were asking me a question !
Now 2. having read through in a bit more detail, I see you are replacing the Matrix power supply as given, with the Steve Bench capacitorless shunt supply. Nice - but you are the local expert on this, I can't answer your question ( my supply is totally different )

[Paul Barker]

I don't see what the dcr had to do with it. The rectifier is not adding DC its passing a half sinewave

Second line from bottom

[Paul Barker]

With utmost respect Mark you don’t say what is the dcr of you’re chokes and what rectifiers used or their dcr requirements, so how can youre anecdote help answer my question?

Well, 1. I didn't realise you were asking me a question !
Now 2. having read through in a bit more detail, I see you are replacing the Matrix power supply as given, with the Steve Bench capacitorless shunt supply. Nice - but you are the local expert on this, I can't answer your question ( my supply is totally different )

I asked a question not specifically you, whoever knows the answer. I think I’ve repeated and further explained , but maybe I wasn’t clear enough that I wasn’t using Stephies Matrix supply , probably what occurred was I’m so knowing what I’m doing I neglected to state what to me is obvious. Probably a feature of dyslexia not phrasing the writen word for normally literate people. If we were in the Owston seminar it wouldn’t have gone the way it had. Oops!

I’m pretty sure the valve data page for the u19 I just pictured for Nick above means the u19 requires dc resistance of 600 ohms in the supply for choke input condition. It is normal for all valve rectifiers to specify such just as it’s also normal to specify a maximum capacitor size as first cap for clc supply even as little as 8 microfarad in cases, maybe even smaller in some.

If I’m wrong about the demands of the data sheet I’d be very happy. I’m all ears. In all cases I’m greatful for the contributions.

[Paul Barker]

If it’s of interest the clc suggestion if I recal without looking further at data, there is a lower resistance requirement for clc. Go figure? I don’t write the data sheet and don’t know why the demands are so, but my kind of unusual mind likes to take the figures printed on data sheet are ignored to the cost of valves in time

[pre65]

As I have a small stash of U19 myself I looked it up.

"The U19 high voltage half wave rectifier was designed to have an effective series resistance of 600 Ohms and to feed a reservoir capacitor of not more than 4 µF."

http://www.r-type.org/exhib/aai0146.htm

EDIT. The Marconi data sheet says 700 Ohms.

https://frank.pocnet.net/sheets/139/u/U19.pdf

[Paul Barker]

Sorry just a last contributory piece of required info. I chose the route I’ve taken because to my unusual mind it’s simpler than the supply in the Matrix. Can you also see that or is it just me? Anyhow with my experience of the r less c less power supply it’s bloody marvellous, which I think I’ve expressed above in the thread.

Furthermore the associated positive voltages are rendered simpler too in my unusual mind. But I’ve in the last few years come to realise I’m not just on the spectrum but severe high functioning autistic formerly known as Sspergers. I don’t mind saying it in this place but I never express it in any other area of life where the negative connotations of labelling mean I keep that to myself.

[Nick]

Yep, but as you yourself have answered that’s the source resistance, so normally provided by the secondary of the transformer, or if required padded out with resistance. In your case its even more questionable if the DCR of the chokes matter. as the 600R will have assumed that there is a cap somewhere that looks like a short to ground. In your case, there is not cap, you have the additional resistance provided by the load and the shunt which should be added to the total resistance.

[Nick]

And looking at your numbers, I would have said that if you have 900v and the total current (shunt + load) is 500ma, thats going to look like 1800R to the rectifiers.

[Paul Barker]

Thanks a lot Nick you are a really clever cookie, I knew you would have the answer, I maybe should have just pm’d you. So I basically misunderstood the diffrence between the shunt valves and the capacitor.

Well that’s mint, I can now wind the chokes for lower dcr, but I’ll have to alter the use of power transformer as my calc of 900v was born from an expectation of 600 ohm before the shunt valve, obviously with 500mA pulling down the voltage I’ll have excess. But is no issue as there are many ways to bring that down at transformer primaries, to arrive at the 900v I employed the simplest option. I’ll just add in the other options which Ive already tested and they work very nicely indeed without the need for the Variac.

As for smoothing capability of these type of supplies, Ive always hit the wallnut with a 14lb hammer rather than a handheld nut cracker, and used every choke to hand, but those were the days when the total demand wouldnt exceed 200mA. Hwen you go for 500mA chokes are few and far between, no cheap ebay chokes at that level.

I’ll wind a couple of 10h at 500mA chokes to add to the one of the same potted double C core aformentioned. If I could buy c cores I’ make it that way but its easy peasy to buy laminations, the clever core materials are out of the reach of a guy doing his own project at home. But in gapped choke talk, I don’t think there is much purpose in niche core material. Brian Sowters dad after all said all the extra inductance of Nickle falls through the gap!

It has struck me however that i the very long term this project is the ideal candidate to finally wind my own 50% Nickel C cores Ive held since long before anyone here ever knew me. I purchased from another diy er in Birmingham who was selling his stuff fater contracting Motor Neurone disease. In those days such items could still be bought in the diy winding community. I had never found a project they were worthy of, since its a waste of Nickel to gap them, and Im not a huge fan of PP, though Im not a stanch enemy of it.

Yes. in the long term, I can employ the Llundahl winding technique but with Nickel C cores, which the Matrix is born for.

Yipee, at last, long after the aquisition of some very special iron/nickel I have a project worthy. I could never find thta type of core material again. The man with MND who even when I bought them in 1996 who couldn’t use his hands any longer and sold his stuff to improve the life of his widdow, so all the money I gave him for iron and Black Gate wkz’s went to a worthwhile home, and his widdow had her life helped for a short time at least. There was noone else there vying to buy his precious materials. I pushed the boat out drove to Birmingham spent what seemed a fortune out of the family budget which I have always had to hide from my wife (its a man thing).

Nick Im overjoyed. in my world of didy tube audio you are second to Stephie! Dont take offence lol ! I was waiting for you to grasp the nettle, I knew you would.

[Paul Barker]

Phil, yes Ive found info on U19 conflicting and uninformative compared to the more common rectifiers. But I supose as they were more for professional use things that didnt need saying in the day remain unsaid, and the guys who commonly used them are now gone. We’ll trail blaze shall we?

[Paul Barker]

And looking at your numbers, I would have said that if you have 900v and the total current (shunt + load) is 500ma, thats going to look like 1800R to the rectifiers.

Wow. How come I don’t know these things?

[Paul Barker]

The beauty of the Matrix is behaves SE but uses pp core methods, slightly uncomfortable with the oarafeed cap on the one transformer, but Stephie said it sounds better than the 845, thats good enough for me. As its only one cap I can ultimately shop for something really special like silver. One stage at a time, at the moment post corona poverty it’s probably industrial polyprop!

Im minted that one day I can employ my Nickel at last.

[Nick]

The only thing I would say is that the 1800R is at your written op points. The shunt is working by "leaning" on the impedance of the supply, so if it wants to bring the voltage down it will draw more current. In this way using a choke input is actually a bad combination as the inherently better regulation of the supply will mean the shunt will have to work harder to bring the voltage down. A higher DCR choke or extra resistor would give it more to work against to give you the required op points.

I guess the lowest the shunt can become in resistance terms is the ra of the valve. And of course the dissipation of the shunt valve is going to be very varied depending on the vr tubes and the mains voltage.

TBH, myself I am with Mark, if the shunt is working why not use a simpler source and forget about it.

[Paul Barker]

Ok, using 3 parallel shunt valves ought to cope and the dcr of the chokes is necessarily high enough probably.