C3g->2A3

What people are working on at the moment
Post Reply
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#61

Post by Lee S »

Had to build a new top plate to accommodate all of the valves. That's taken me 4 hours. Then I have had to strip down the LW and clean all the bits up for re-working. I have just fitted the valve holders to the top plates (after realising that I had no little nuts and bolts, so had to use screws and computer stand-off pillars... rather gash, but will do for now). Just finished all that crap, so now I can actually begin wiring it all up... Sheesh... I will be back in a couple of hours..... or so!!! :roll:
©2020 Lee
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#62

Post by Lee S »

OK... It's up and running after a load of hassle with the chassis. First off, it seems to work OK, but there seem to be a couple of problems.

The gain seems to be quite low in comparison to the Loftin White. I need to give it at least a third of a turn more on the volume control to run at the same volumes. I am not sure if this is anything to do with the second problem detailed below.

I get 220v across the 0A2 all the time. It's as though it's not firing up. Removing it whilst running has no effect on sound, or voltage. It's as though it's not even in. I will double check the connections in the morning. Weird. The only thing I have changed from the circuit diagram is the grid-stopper on g2 of the C3g. I used a 1.5k instead of a 1k. Wouldn't matter would it?

Sounds OK though. I guess it needs to run in for a few days. I will report back about its sound when I have sorted out the problems and let it settle down. Initial thoughts are that its a little more refined and smoother than the LW.

Cheers
©2020 Lee
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#63

Post by Lee S »

I guess the VRs would never work without a cathode connection!!! Doh!! :oops:

Well that's sorted out now. Nice pink glow from the VRs and 150v across the C3g's g2. Gain is still a little low though. Maybe it's the 1M grid leak resistor? Maybe it's the 100k pot I am using rather than the 250k James specified? Any thoughts?

Yours stupidly,
©2020 Lee
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#64

Post by Nick »

What was at the front of the LW (valve) and what anode resistor is on the c3g?

I guess the other question is whats the problem with having the vol control a bit further up?
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
simon
Thermionic Monk Status
Posts: 5600
Joined: Thu May 24, 2007 11:22 am
Location: People's Republic of South Yorkshire

#65

Post by simon »

I have mine with a 100k pot and no grid leak resistor at the moment. Wouldn't say it made that much difference to volume.

I seem to remember the LW did go loud for not a lot of pot turn. I suspect that James has designed the amp so that all the pot is usable - Steve S has written a fair bit about this recently on here.
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#66

Post by Lee S »

There is nothing wrong with turning the knob up a little. In fact, it is probably better. I was probably a bit rash describing the gain earlier. I think I need the volume at about twelve o'clock instead of ten o'clock. Its just different I think. Seems OK though and goes loud enough. A problem that is not really a problem me thinks.

The LW used a 100k log straight onto the grids of a paralleled ECC83 Nick. Now using the same 100k log onto g1 of the C3g (with a 1M grid leak in between). The anode resistor on the C3g is a 7.5k and the cathode is a 75R.

I was a bit tired and hacked off last night to get a proper impression of it. I am returning now with fresh ears. It is definitely no worse than the LW, that's for sure. It seems a little less "rough around the edges" and more refined, but still retaining the same drive and authority. It also seems a little less compressed and more open sounding than the LW. I think the refinement appears to make it look less exciting, but in fact, when you listen, it is still doing everything well, so this is good. It means I won't tire or get bored of it.

Players and instruments seem to have a little more air and space around them in the sound stage, making for a slightly bigger, deeper but more importantly, a more believable sound stage. Tone seems very much unchanged. No extra highs or no extra bass. Mid range is faultless as usual. Bass seems a touch tauter and detailed and the highs are all there, but without any shrillness or glare. It has plenty of pace and doesn't sound slow or too laid back, but not too forward either. I am liking it.

I will live with it "as is" for a few weeks and see how things go. I am using an Alps Blue 100k pseudo-shunt pot with 80k series metal films. All other resistors are either 2w or 5w Kiwame with 12W Mills for the 2A3 cathode resistors. Hammond 10H 65mA chokes in audio circuit. 100uF Black Gate cathode by-pass, 0.47uF Jupiter coupling caps, 94uF Rifa C3g bypass cap, 22uF Solen ultra-path cap and James 6123 output transformers on 3.5k/8R. The PSU is using all Hammond iron and Mundorf caps. Voltages are all OK and nothing is amiss.

A strange phenomenon I get is a loud-ish pop through the speakers at switch on, when the VRs fire up and begin working. Can anything be done about that?

I have a list of small tweaks to try in a few weeks when I am used to the sound. Any comments or any more to add? Add snubbers to both of the PSU chokes. Remove the 1M grid leak resistors. Replace the 2 x 47uF Rifa caps with a very large (or a couple of large) film caps.

Cheers
©2020 Lee
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#67

Post by Nick »

I don't think the grid leak will be doing any harm, and stops the grid floating if the pot wiper leavs the track.

With a 7k5 load the gain should be aprox 100 (gm * rp as an aproximation), so I would have expected if anything it to be a little higher gain than a ecc83. What grid leak is there on the 2a3?

Assuming the VR cathode current is going through the cathode resistor as well as the pentode current, I would try with that unbipasssed to see if you can hear any difference.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#68

Post by Lee S »

Hi Nick. C3g cathode is un-bypassed as it is. The 2A3 grid leak is 220k. I will leave the C3g's 1M grid leak in if it does nothing to the sound.

Cheers
©2020 Lee
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#69

Post by Nick »

Ok, ok, it was the
94uF Rifa C3g bypass cap
That made me think it was bipassed. Anyhow, 220k on the 2a3 isn;t going to alter the gain of the c3g much, maybe make it 95. But as there is enough gain its not a big deal. What current are you running the c3g at, maybe the Gm is less than the 14 I used to calculate the gain.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#70

Post by Lee S »

Sorry Nick... my bad. The 94uF is not a bypass cap, but more of a PSU decoupling cap. It goes from the top of the C3g's anode resistor to ground.

Not sure what the current through the C3g is to be honest. I will have to get the meter out and have a probe around and get back to you. Give me an hour or so.

Cheers
©2020 Lee
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#71

Post by Nick »

Its not a big deal, I was just wondering why the gain was a bit less than theory suggested.

Always worth measuring a couple of voltages though to see things are behaving as you expect.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#72

Post by Lee S »

Hi sorry for the delay. The voltage at the 7.5k resistor at the top of the C3g circuit is dead on 330v with 220v beneath it at the anode. g1 is at 0v, g2 is at 150v so I am guessing the valve is operating near the specified values, so about 15mA or thereabouts. How would I calculate the current from the voltages above? It's surely not just a case of adding the anode load, cathode load and plate resistance together and using ohms law? Not sure what the plate resistance of a C3g is. The data sheets are all in German.
©2020 Lee
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#73

Post by Nick »

Yep, just ohms law, 330-220 = 110v, 110v across 7k5 = 14.6ma, ignore plate resistance,as that sthe dynamic resistance, and should be very high for a pentode. Look at the slope of the grid line at the op point to get an approximation of the expected rp. The voltage on the cathode will give you the total current at that point, you can find the current into the top of the VR tube using the B+ - (150v + Vc) to see the voltage across the resistor to the VR tube, that gives you the current down that leg. If you have a grid stopper before g2, measure the voltage across that, that will give you the current into g2, then the current into the VR tube, is the current into the top, minus the g2 current. Then the anode current + the g2 current + the VR current should equal the current through the cathode resistor, so you can pin down all the voltages and currents around the valve.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Lee S
Old Hand
Posts: 664
Joined: Fri Aug 10, 2007 9:39 am
Location: UK

#74

Post by Lee S »

Of course.. Ohm's Law and Kirchoff's Law. Cheers Nick. So what is the relationship between the anode load and the cathode load with respect to the C3g circuit as a whole? The spec sheets say to use a 115R cathode resistor but James specced a 75R. Why? Does Rk control the current flow through the valve?

Anyway, it looks like the circuit is operating spot on as advertised. I must say that it's sounding rather nice now. It's opened out a touch too. It was doing all sorts of funny things this morning. First it sounded good, then crap, then mediocre and now it's sounding really good.

I notice that when I turn the volume control all the way down too, I get a strange kind of buzzing pulsating noise. It goes when you dial in just a touch of volume. Does this mean that I should fit a grid stopper to g1?

Cheers
©2020 Lee
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#75

Post by Andrew »

Here's one in English....

Dam bigger than 256k, can I e-mail you it Nick?

cheers,

-- Andrew
Post Reply