audio-talk

On the GK-71 thread in Valves I set out two operating conditions for the beastie for 12Watts and 25Watts output along with some example circuits. I then foolishly ordered two pairs of the beasties

So now I've started costing the build for the 12W design and decided it was getting too pricy for me at present with everything else I'm doing - not least getting the AERs repaired! What to do? Well, whilst investigating costs for OPTs I realised that the Hammond 125ESE is about as cheap as one can go and still have good power (please let me know if their are better or cheaper alternatives!) so I've based a design around that OPT. That sets a limit of 80mA on anode current and looking for a good op point for that I came up with the following conditions for the GK-71:

Load Line 5000 R
Power 9 Watts

Va 500 V
Ia 80 mA
Vg -72 V
Pa 40 Watts

Rk 900 R
P(Rk) 5.76 W
Ck 35 uF
F3 5 Hz

That looks good on paper - still 9watts out

The simplest circuit is the two valve one but on further analysis using the PL802 as the first valve has some problems associated with it - namely excessive millar capacitance that requires a 100R driving impedance - Ouch

I has flippantly remark at Owston that driving the GK-71 with 6AM4 would be good as this is a tiny, tiny valve designed for VHF/UHF use. However this is just what we need as it ensures low millar capacitance with high gain. But is it beefy enough? It will run 10mA anode current and that is marginal for 72Volts swing into the GK-71 grid but on the Quasars I am unlikely to run 9watts so I'll chance it - we'll see what it sounds like in the room at Owston!

6AM4 Operating Conditions
Va 200 V
Ia 10 mA/V
Pa 2 W
Vg -1 V

Rk 100 R
Ck 318 uF
F3 5

Rload 38000 R
Av 69
P(Rload) 3.8 W
B+ 580 V

Right on the max levels for a 6AM4 but its at least 5,000 hours at this level and they are cheap - at present anyway

The circuit is attached. I'm using Andrew Is filament supply idea (thanks Andrew for the inspiration!). The 110uF psu caps are series connected 220uF at 350V as these are very cheap! I haven't found a good source for the mains trannie yet that is cheap. If necessary I'll use a 230:110 transformer in reverse - not a good idea if other options exist but doable until the budget increases...

Now just to get all the parts and build the damn thing!

James

That looks good.

I have a dual mono PSU that might just stretch to 500v, and I also have my 5K "big" Hammond OPT.

Tempting.

Excellent James! Funnily enough that's almost exactly my design too, though I was going to run the GK71 just a smidge hotter at 95mA. Need to check my calcs though as I was only getting about 7.5W. Where did you get the GK71s? Cheapest I saw on Ebay worked out at 20 quid each, same as a GM70.

Funnily enough after you mentioned 6AM4s I checked when I got home and I have a few. Cost the princely sum of a dollar each (plus shipping and taxes) a couple of months ago .

I paid £18.00 for a pair and £23.0 for the second pair plus shipping. I had to search and grab them quickly for these prices as most were that much for one valve, as you found.

6AM4 is a sleeper - I have a half dozen and bought them to use in a phono that I never got around to building... but don't tell anyone how good it is - I have used them as a voltage amplifier before and they sound nice.

J

Nice use of the CMC and chokes to clean up the SMPS - and also getting 'two for one' assuming indirect heating on the driver.

Do we have any feel for where grid current starts on the GK-71, or the GM70, as I'm definitely hoping to use an A1 driver on mine. Perhaps one of the uTracer guys can help us ....

I've done some research and not come up with a lot. The datasheet on the beastie has a plot of grid current against grid volts and it passes right through 0V and 0mA leading one to suppose that grid current doesn't flow early i.e. before )v on the grid. This seems to be bourne out by the anecdotal evidence of comments on various forums that it is easy to drive... Thats all I have - I hope someone has more info!

Andrew Is SMPS use is a great economy idea and the clean-up parts are in my bin already so daft not to use them!

James

Please excuse my ignorance but....what are the Tx/choke things in the heater circuit?

Hi Ed,

the first choke with the windings on both sides is a common mode choke and its used to remove signals that are common to both sides of the choke - these are typically rf moise and hash that was mains borne but with a SMPS might be noise at the switching frequency of the supply. Here is wikipedia take on it http://en.wikipedia.org/wiki/Choke_(ele ... mode_choke

the second chokes are a conventional pair of chokes and these are there for two reasons 1) to present a high impedance to the audio signal in the filament so it doesn't flow into the filament supply but let the dc filament heating current through unimpeded into the filament and 2) to prevent any differential noise from the SMPS getting into the filament.

J.

thanks James, super explanation.

JamesD wrote:Hi Ed,

the first choke with the windings on both sides is a common mode choke and its used to remove signals that are common to both sides of the choke - these are typically rf moise and hash that was mains borne but with a SMPS might be noise at the switching frequency of the supply. Here is wikipedia take on it http://en.wikipedia.org/wiki/Choke_(ele ... mode_choke

the second chokes are a conventional pair of chokes and these are there for two reasons 1) to present a high impedance to the audio signal in the filament so it doesn't flow into the filament supply but let the dc filament heating current through unimpeded into the filament and 2) to prevent any differential noise from the SMPS getting into the filament.

J.

hello all.. i looked at that curcuit, that looks do-able i thought. haveing just read that not so shure but we will see, and yes im prob one off the people pushing the cost of GK-71 up i have 3 on there way

I was going to use a couple of damper diodes for rectification but thinking again I can see the benefit of using SS diodes as they'd save 30-odd volts of HT. But I like the thought of a soft start to the HT - am I just conforming to myth and is there nothing really to worry about with instant HT?

I switch my 833 HT straight on with only an inrush limiter to slow things down slightly. It doesn't seem to do any harm! (howls of maniacal laughter) The GM70, you see, has damper diodes for that oh so easy smootch into HT.

I wouldn't say it was myth but whether it is applicable advice to us is another question

The advice, from the time that valves were all there was, to delay the switch on comes mainly from two places - domestic valve TV sets and from big valve transmitters - in both cases there were good reasons for it that generally do not apply to us and have the same root cause - multi kilovolt B+, in this case the very strong electric fields pulling on the filaments and grids before they have heated and the space charge established in the valve could cause damage to the valve either by distorting the grids or by stripping the cathode of free electrons and possibly of some of the filament coating (where it was used).

Our amplifiers do not run multi-kilovolt B+ so its not a problem for us to really worry about...

I find a ss bridge, when properly snubbed, usually has better bass response than a valve rectifier and that is my main reason for using them

J

It's a joy (to me) to see HT ramp up gently after giving the heaters time to warm up a bit.

Is it essential - probably not.

I also use inrush limiters.

Nice explanation James, that makes sense.

James wrote:I find a ss bridge, when properly snubbed, usually has better bass response than a valve rectifier and that is my main reason for using them

I've seen ss rectification for a stiffer supply with diodes in the earth connection before, jc morrison I think? Are there any guidelines for determining the correct snubbing - do ultra fast diodes help?