C3g->2A3

[Lee S]

Hi all.

I have decided to build another 2A3 based amplifier, based on James D's C3g/0A2/2A3 design that Simon has also built, maybe changing it later on to run on 45s. Simon has kindly given me the schematic albeit minus a few values and voltages. I have a few questions about this amp if you kind gents have time to answer them.

My main concern is the PSU. The mains transformer in the schematic has no values attached. I need to get 325v B+. The setup is GZ34->CLCLC. I don't really want to buy another mains trafo if I can help it, but the one I have (Hammond 374BX) is 375-0-375. I have been fiddling with PSUD and according to that, if I change the first cap for a RC stage so I have RCLCLC with 47R->2.2uF->10H->100uF->10H->100uF, that should give me about 325 ish volts at about 170mA (is this about right for the circuit?). Is this OK? I am using the Hammond 193J choke already (10H 200mA), so I can just get another one. Is a first cap this low OK?

Any idea what spec the 10H choke in the HT line for the C3g is? I am guessing at about 10H 65mA or 100mA will do, preferably the 65mA. Any thoughts?

Also, any thoughts on resistor wattage? I will get 12W Mills for the 2A3/45 cathodes, but I am guessing that the rest would be fine at 2-5w?

PS: Anyone got any spare 0A2s they don't need? Cash or swaps.

Cheers

[Nick]

It might be better reducing the 2.2uf at the front instead of adding the resistor.

[Lee S]

Cheers Nick.

I can remove the resistor in PSUD and it makes very little difference to the voltage actually. I will play around with somewhere between 1.8uF and 2.2uF. I think I will get a 2.2uF to start with, to get it up and running and see what voltages I get.

Thanks

[Lee S]

How crucial are the de-coupling capacitor values? I have a couple of 22uF Solens that will do for the B+ -> 2A3 cathode. Would a 47uF/450v Rifa work for the C3g de-coupling cap? I know it's over twice the value, but it's the closest I have at the moment. Don't want to shell out on caps yet until I have heard the amp playing. I guess it would work without the C3g de-coupler, so I am also guessing that 47uF will be good enough to get it up and running?

Cheers

[andrew Ivimey]

almost anything will be fine to get it working and ..... when it is, I bet you can't hear the difference between 47 and 22mfd.

this is presumably the cap that is across the HT of the C3g ???

this is a 'de-coupler'? I can see how it helps to provide a sort of reservoir for the C3 which, oh okay, sort of gives it its own voltage - well I never!

Onwards!

I have even been cutting drilling and bolting the chassis for 'Amity' but it is oh so cold.

[Lee S]

Yeah... This is exactly what I thought Andrew. I have a couple of 47uF Rifas that will do. If it sounds nice they can stay in there. I thought it was a decoupler, but as you say, it's probably some kind of extra smoothing or filtering after the choke.

Onwards AND upwards indeed.

Good luck with the Amity. Is there a schematic for this beauty? PP isn't it?

Cheers

[simon]

Give it a whirl Lee, it's an interesting alternative to the Loftin White.

There's quite a few points here.

James designed the amp based on the bits I had in the LW from Philip. So the good news is that the 374BX mains tx is exactly the one I use.

The circuit shows a GZ34 but I had a few problems with modern GZ34s so I stuck a Mullard GZ37 in and forgot about it . The small first cap is fine - it's just to trim the HT in the choke loaded PSU. I ended up with a first cap of 2.2uF, but it'll depend on the DCR of the chokes you use. 2.2uF should be close enough to start I'd guess. James reckoned the amp would work fine with HT between 275V and 375V. The first cap needs to be 630V or greater.

When I get chance I'll try the PS with just 1 choke (tweaking the first cap accordingly) and see if it makes any difference to the sound. If you want to minimise spend try a single choke first, and if there's too much hum maybe try a second R stage?

James specced the chokes in the audio circuit at 40mA.

I used 1W resistors generally IIRC. However, the 7.5k C3g anode resistor needs to be 4W, and the 15k on top of the 0A2 needs to be 5W and ideally 10W. The 12W Mills should be nice - I have ordinary 14W enamel wirewounds. I haven't used hum pots - I used 2 No. 10R resistors instead and the amp is pretty much silent. There's also a 100R or 150R on G2 of the C3g too. Gridstoppers aren't shown but were definitely needed - I had some interesting side effects until Nick sorted it out!

The 22uF caps - James recommended films here and I used Solens. I have a 22uF for the 2A3 but I found that I needed a much bigger cap next to the C3g - I ended up with about 60uF I think. Without the extra capacitance I had lots of impressive speaker cone oscillation. Another thanks to Nick.

Try the 'lytics, see what you think. Discussing with James he felt the difference might be due to his version being in a metal chassis and mine on a breadboard...

I have slightly bigger coupling caps too - 1uF.

I'm sure I've got a couple of spare 0A2s somewhere I can send you - I'll have a look.

Trying 45s is dead easy. Change the cathode resistors to 1k5, and retune the PS - I changed the first trimming cap to 1.5uF. You can use different OPTs if you have them, or 4R taps instead of 8R, say, with 2.5k OPTs. The driver stage is fine as is.

Good luck, and you're definitely on the slippery slope now .

[Lee S]

Hi Simon.

Excellent information. Just what I was after. Many, many thanks. The slope is indeed getting rather slippery and a little steeper every day, he he!!

I have ordered most of the bits for this amp tonight, so hopefully I should have all I need within the week apart from the 0A2s and the audio circuit chokes. I will have a poke around on eBay before giving Mr Ramsay a call. I got a 2.2uF 800v cap for the first PSU cap, so that should be good. The 47uF Rifas should work for now until I can get some bigger ones. I got 5w for the C3g anode and 0A2 anode. Maybe I will get a bigger one if it shows signs of overheating. Should be OK to get me going. Rest are 2w or 3w. Should work OK. I will get some 100R gridstoppers too. Need to pop to Maplins tomorrow anyway so I will stock up on some and get the 100nF caps for the 0A2 as well.

I wasn't going to bother with any kind of hum circuitry on the 2A3. Was going to leave it as in the LW, with the cathode resistor and bypass cap connected to the centre tap of the filament trafos. I have no hum at all on the LW.

What did you think of it sound wise Simon? Was it as good as the LW? I like the LW sound and don't really want to take it apart, so I may just get another couple of UX4 sockets and build a new breadboard with the new one on and then just swap between them. Have you tried it with 45s?

Thanks again for the info. Much appreciated.

Cheers

[david C]

definitely on the slippery slope then Lee

[Lee S]

It looks like David.

I will have a fiddle with this circuit of James' and maybe try the C3g in triode mode too. Then maybe replace the 2A3s with some 45s, maybe CCS on the output valves and finally, change the driver circuit to something different (unless it sounds to my liking).

I was going to have a bash with PT15s too. I might try and get a pair to play with, but I don't want to shell out another load of cash on a pair of 5-10k OPTs, so it would have to be done with the James trafos. Not ideal as they probably don't have enough inductance even on 5k:4R config.

The DL103 should be turning up either Wed or Thu so that will distract me for a bit whilst these bits turn up. Should be up and running with the amp in a couple of weeks. I may build a Baby Ongaku or Bugle if I don't fancy this one. We'll see.

I know Philip (Pre65) built James' C3g->PT15 amp. I wonder if he had any problems with grid stoppers, filter caps, etc on the driver stage of his amp? I think it was exactly the same driver stage.

Cheers all

[Clive]

Well Lee, it seems like you're just about the at the point of no return

Re the centre tapped heat tx / cathode connection, I found on David's Bugle that with one set of 45's the hum level was very low, with another set the hum was a lot higher. It all depends on pot luck as to how symmetrical the heater winding is. So I can run other valves I've just installed humbuckers. I'll have to try some TJ45s as I'm sure the old 45s I have here must be well past their best.

Cheers,

Clive

[pre65]

Hi Lee-it did confuse me that the gridstoppers were not on the original circuit.

I built it with 150R carbons,but at Witham Andrew felt it needed a bit more so i have got some 1K carbons to try (soon !).

I used DC for the C3g and regulated DC for the PT15,but if i was starting again i would try AC first.

As Simon says the 22uf cap is too small,i used a 100uf in that position.

The cap marked 22-100uf i used an 80uf motor run cap and use motor run caps in the PSU,which was based on the Rankin 45 dc cc circuit.

The input 250K and 1M resistors were omitted and a 220K put in their place.

Again,if i was starting again i would use those 6D22s television damper diodes (2 off) instead of a full wave valve rectifier (like wot i did on ECL82 Doddington special )

Good luck wiv yorn !

[Lee S]

Cheers guys.

I'm not going to bother with DC anywhere if I can help it. More hassle than it's probably worth at the moment. I just want to get a general taste of the amplifier's character first. I have a couple of 100uF caps somewhere too. I will try those if 47uF is not enough.

I am not going to use a 250k pot on the input, but a 100k, but I will keep the 1M grid resistor in and try it like that.

Philip. When you say you used a 80uF run cap, was this instead of the 100uF cathode bypass cap or the 22uF WE/ultrapath cap? I have a 100uF/100v BG for the bypass and a Solen 22uF for the WE cap. Should do shouldn't it? Simon used a 22uF here.

Clive. I know what you mean, but again, it will be a case of throwing it together and then adding one if it does hum with any of the valves I have got. The 2 sets of 2A3s in the LW were both as silent as church-mice. The new pair cured the rustle I had too.

Cheers

[pre65]

Hi-on "my" circuit there is a cap between the 10H choke and the 22uf ultrapath cap (and a 2.88K resistor)

See the circuit that "Pedro" from the WD forum sent me (the James D one)

[Lee S]

Ahh... the 2A3 circuit doesn't have these caps and resistors. Cheers Philip.

I wonder, when the driver stage is identical, that the CR shunt components on the 0A2 are different, as well as the 0A2's anode resistor? Mine shows a 15k on the anode with 390R/100nF CR components. Yours shows 12k anode with 100R/300nF. What do these components actually do? Some kind of filter or zobel or what and why the different anode resistance? Does it alter the regulation?