6J5G --> 2A3

Lee S · Post by **Lee S** » Thu Apr 17, 2008 10:17 pm

How about this guys? Would it work?

Could I drive the 2A3 from a single 6J5G, or would I need a driver valve too?

I fancy trying a triode front end instead of the pentode and also fancy playing with plate and grid chokes on the front end and a grid choke on the 2A3 too.

Saw this and thought it may work with a 2A3 instead of a PX25 :-

http://lundahl.se/img/armada/6j5-px25_schema_oct04.gif

Any opinions?

Would it be better with two 6J5Gs or a single 6SN7? Is the 6SN7 valve crap or is it not bad?

How about 6J5G SRPP?

Could I use grid and plate chokes on a pentode such as the C3g?

Are there any better triodes I can use than the 6J5G?

Cheers

Post by **Nick** » Thu Apr 17, 2008 11:44 pm

Err, I will try and answer in order:

Yes

Yes, maybe, best trying it and deciding

Should work, may sound good, the auther seems to know what he is doing.

Don't know, I haven't compaired two to one, its not crap, but there may be better,

Yes, thats another option.

No, plate chokes are a bad thing on a pentode.

Yes.

IMHO, read less internet, build more on breadboard. Trying to find the best solution before building will fail, and might miss the perfect solution for you, as its not what worked for someone else.

And as always the above is IMHO.

Lee S · Post by **Lee S** » Fri Apr 18, 2008 6:50 am

Cheers Nick.

I will purchase a couple of octal bases and some 6J5Gs and have a play first. Then I will get a couple of grid chokes, which I can use on any amp I guess, so that won't be money wasted. Then I may have a play with plate chokes and possibly ITs later on.

Thanks for the advice and answers,

Cheers

andrew Ivimey · Post by **andrew Ivimey** » Fri Apr 18, 2008 7:53 am

Lee are you gong to Owston? I'll bring some 6J5s for you.

al newall · Post by **al newall** » Fri Apr 18, 2008 8:32 am

Feel free to use me as a "go between", if Lee has any plans to come over to Chester in the near future.

Lee S · Post by **Lee S** » Fri Apr 18, 2008 12:31 pm

Sorry guys. Can't make it to Owston this time.

Andrew, many thanks. I will take you up on the offer, just for a try. If I like, I will buy some in to keep. I will fit the breadboard with four octals and try a few different configurations. That will make it easier. I can also try some different octal valves up front too. Hand them over to Al and I will pop over and get them next weekend. Cheers,

Al, I am back in Cardiff now, but could pop over Sunday evening (27th) on my way back down if that's OK?

I am thinking that the 6J5 -> PX25 on the link I posted is a power amp designed for use with a pre-amp as there is no pot on the front. As mine needs to be integrated, I am also guessing that I am going to need a bit more gain. This is why I thought of using two 6J5s. A bit like this:

http://www.vt4c.com/bb/files/6sn7___208 ... a3_121.jpg

Would a single 6J5 and a 2A3 give me enough gain? I think it is only half the gain of my C3g I am using at the moment.

Thanks fellas.

Cheers

al newall · Post by **al newall** » Sat Apr 19, 2008 6:15 am

Should be ok Lee, but i'll need to check tomorrow and let you know.

Lee S · Post by **Lee S** » Sat Apr 19, 2008 12:33 pm

OK Al. Many thanks. It will literally be a 5 minute flying visit unfortunately.

On a different note, looking at the circuit diagram for the Sun Audio 6SN7 -> 2A3 amp, the two halves of the 6SN7 are directly coupled. This may sound like a stupid question, but why are some stages directly coupled and some need capacitor coupling? This is something I am not sure of.

I can obviously use a single 6J5, 6J5G, 2C22, blah, blah, blah instead of half a 6SN7, but why doesn't each stage of the 6SN7 have a grid leak resistor?

Is this a poor circuit?

Have a good time at Owston guys!!

Cheers

david C · Post by **david C** » Sat Apr 19, 2008 12:42 pm

it's interesting, these are questions that cross my mind but I'm too afraid to ask

Post by **Nick** » Sun Apr 20, 2008 5:37 pm

Well, the simple and short answer is that direct coupling is used because it avoids a cap. Sometimes its a good thing, sometimes I wonder if it creates more problesm than the cap it replaces. The two stage 6sn7 direct coupled setup is a standard driver stage that seems to have been about forever.

The dpwn side (IMHO) is you are forced into op points for the valve that allows the direct coupling, and that bit I wonder if its worst that the cap it sreplacing. I guess it saves on the cost of a cap, and the sound of that cap will be heard, so its possible a fair saving of cash.

The is so great leak, as the grid gets its voltage reference directly from the previous anode. The grid leak is needed when there is a cap before the grid, as then there is no DC path to ground (or B+) which are equivilant in this regard.

Lee S · Post by **Lee S** » Sun May 25, 2008 12:37 pm

Thanks for that Nick. I have read and reread this and I think I am understanding now.

So in a directly coupled two stage, the first stage's anode voltage is obviously seen on the grid of the second stage valve. Setting the cathode of the second stage valve to give the recommended g1 voltage then balances things out. So if the g1 voltage needs to be, say, -12v, and the anode voltage of the first valve is 60v then the cathode of the second valve needs to be at 72v?

Does this mean that the operation point of the first valve could potentially be running "cold" so as not to put too high a voltage on the second stage's grid? What are the downsides of running a valve that, say, in the spec sheet, lists the lowest anode voltage of class A as 90v, at 60 or 70 volts instead? Does it harm anything?

I suppose it's all about finding the correct points. If I ran the first valve at 200v, then there would be 200v on the g1 of the second valve, which, to be biased correctly, would need its cathode to be raised to about 210v which, would then give very low voltage swing as the maximum anode voltage could be potentially only be 250v, so giving just a 40v swing. Whereas if I used just 60v on the anode of the first stage, and ran the second stage, directly coupled, at Va 250v and Vk 72v (to give me the -12v g1) then I would have 178v swing. Is this correct?

Also, if I tie the grid of a R-C stage to ground via a grid leak, and the cathode of that valve is at, say, 50v, then g1 would be at 0v (ground), does this mean that the bias is -50v for that stage then?

Cheers,

Cressy Snr · Post by **Cressy Snr** » Sun May 25, 2008 1:07 pm

Lee S wrote: Does this mean that the operation point of the first valve could potentially be running "cold" so as not to put too high a voltage on the second stage's grid? What are the downsides of running a valve that, say, in the spec sheet, lists the lowest anode voltage of class A as 90v, at 60 or 70 volts instead? Does it harm anything?

Cheers,

Hi Lee.

The problem of cold bias on the driver stage is a very real one and limits the choices of driver valve available in a direct-coupled circuit, unless you want to have crazy voltages at the anode/cathode of your output valve.

In my new breadboarded Loftin-White PX25 amp, I use the "unfashionable" ecc88 type (actually a Siemens CCa in my amp) as the input/driver valve, precisely because it is happiest, and at its most linear when working at around 90Va. This fact makes it ideal for direct coupling as it can also be made to pump out serious current with minimal effort on the part of the builder. Choke load it, feed it from a low voltage, SS rectified choke input supply putting out 100V and it sounds superb. It drives my PX25s beautifully.

The high value cathode resistors I use are a 100W composite of 4K7 ally clad wirewounds. These generate a cathode voltage of around 120V, 20V above the grid voltage of 100V. My PXs are dissipating around 14W so they are running at a bit over half power, with 280 odd volts across them.
Nice and easy.

Steve

audio-talk

6J5G --> 2A3

#1 6J5G --> 2A3

#2

#3

#4

#5

#6

#7

#8

#9

#10

#11

#12