Fender Super Reverb - potential volume problem

[IggyRoyale]

All,

As my homebrew Super Reverb journey is nearing the end, I'm trying to get rid of as many bugs before I can start using this amp live.

I have a question with regards to the output volume. This is meant to be a 50W amp (with two 6L6 GC tubes in the power stage). Now, I'm not a volume freak as I mainly play in small venues. I've heared and played through 50W amps in the past and know that my amp shold be louder than it is. I cannot get any decent volume out of it unless I'm using a pedal or an FX board. It is even worse if I'm playing my Strat (single coil). I have to turn the master volume up to about 9 to get it to come to a "nice" level.

Are Super Reverbs known to be "quiet"? Either way, can anyone suggest the reason for this? I suspect a few things, some of them being wrong resistor values in the preamp, output transformer problem etc. Where do I start?

Thanks
Iggy

[Enzo]

They are not known to be quiet, but keep in mind you don't have a Super Reverb, you have something like one that you built yourself. We can usually assume the parts are correct in a factory amp. When we make the amp ourselves there is no guarantee the parts are corect. A 470 ohm or 4.7k ohm grid resistor where a 470k ohm ought to be will quiet any amp.

Amps are all the same under the skin, is this really much more than a Bassman with effects? Start? I'd start by going tube to tube in the opreamp checking plate voltages and cathode voltages looking for some out of range. I'd be verifying each and every resistor and cap for value. I would make sure the screens on both power tubes had B+ and I would check that the bias on the power tube grids was not sky high. I would apply a steady signal to the input and scope it through teh amp, or at least follow with an AC voltmeter to see where it diminishes.

[IggyRoyale]

Thanks for your input. Had another thought ... considering that there is no unexpected volume diference between the two channels the problem might be in the power stage. I will check all voltages on the power tubes. I did check the bias on the power tube grids ... together they appear to be drawing just under 60 mA.

I like the suggestion to follow voltages along the signal path. I assume I'd have to measure them to the ground? Would you like me to send you voltage redouts per tube/pin?

Cheers

[Enzo]

I did check the bias on the power tube grids ... together they appear to be drawing just under 60 mA.

This sentence potentially scares me. How did you measure the 60ma? You sure we are not talking 60 VOLTS on the grids now? That ought to suck some life out of them if so.

[IggyRoyale]

Enzo,

I've used cathode resistor method to bias the power tubes. I've installed a 1 Ohm current sensing, cathode resistor and measured voltage drop across it. I measured 28 mV which, according to the text book, translates into 28mA (http://www.geofex.com/tubeampfaq/TUBEFAQ.htm#howtobias). Considering that there are two power tubes in the circuit 28 x 2 = 56 ... in other words just under 60 mA as I quoted above. Am I misinterpreting something here? Is this the right current draw for two 6L6GC WC tubes?

Either way, I re-checked all resistor values in the preamp section and I did find a wrong resistor in the reverb circuit. Replacing that to the correct value improved the output volume immensely.

Another question though, the tube I am using in the phase inverter circuit is 5571 rather than 12AX7. Does this matter? Can I swap it to 12AX7 without modifyung the circuit? Would it make any difference to the overal output volume?

Keen to hear your further comments on the tube bias ...

Many thanks for your input so far.

Iggy

[MWJB]

You would normally express bias current as current per tube, e.g. 28mA...which is ballpark for a Super Reverb. Most people would have assumed that the "60mA" you mentioned was for one tube...too much in a SR.

The phase inverter tube in a Super Reverb is a 12AT7, but you can try 5751, 12AX7, 12AY7, 12AU7. The 12AX7 will decrease headroom and thicken the tone, the stock 12AT7 was fitted to clean up the tone and makes the amp "chimey-er". 12AY7 or 12AU7 will increase headroom further but your tone may go a bit "sterile".

You could even try the 5751 in V1, or V2 (depending on whether you use the normal or Vibrato channel) with a 12AX7 in the phase inverter.

Use whatever sounds best to you, though most people love them just the way Fender intended. I'd always keep a 12AT7 in the reverb driver (V3) and a good 12AX7in the reverb recovery (V4).

[Enzo]

Um...

You have cathode bias, that is fine. Is it a common cathode resistor? That is, it sounds like you have the two tubes running through a single resistor. Then you added a sungle 1 ohm resistor as a sense test point. If that is the case, you don't have 28ma x2, you have 28ma TOTAL, which is 14ma per tube. If each power tube has its own cathode resistor and each one its own 1 ohm resustor, then your 28ma would be correct.

[IggyRoyale]

Enzo, MWJB,

To make this discussion easier, here's a snippet from my power supply wiring diagram. It shows where the current sensing resistor is. Now that I think about it, yes I stand corrected. In this setup measruement would say how much current the two tubes draw together i.e. 28 mA or 14mA each. Can you please suggest what the right bias reading should be?

Kind regards

Iggy

[Bruce / Mission Amps]

Enzo, MWJB,

To make this discussion easier, here's a snippet from my power supply wiring diagram. It shows where the current sensing resistor is. Now that I think about it, yes I stand corrected. In this setup measruement would say how much current the two tubes draw together i.e. 28 mA or 14mA each. Can you please suggest what the right bias reading should be?

Kind regards

Iggy

Not picking on you, but do you know what you are doing there with the idle current info?
Depending on your plate voltage you should be running these individual tubes over 28ma to sound halfway right.
With your setup, (one current sensing resistor) you want something closer to 56ma-64ma... you'll be reading that will in mv dc with your test setup.
You probably should change that so you can see what each tube is doing.
I run them hotter then 28ma each but I always measure the plate voltage first and keep checking that as I increase the idle current until the 6L6s are running at least about 14-15 watts each.
Also... ... why are you running an 820 ohm FB resistor on the 8 ohm tap?
Do you understand that voltage divider and why 820 is incorrect for an 8 ohm secondary and load?
That's sticking way too much NFB voltage back into the PI for this amp.
Change that to a 2700 ohm to 3900 ohm at let that PI breathe a little.

[Enzo]

I'm sorry, you don't have cathode bias then, just a cathode sense resistor. Above comments still valid, you picture clued me in to what you built.

[IggyRoyale]

Bruce,

Thanks for jumping in on the thread. You are quite right, I don't fully understand the workings of this circuit. I am learning as I go. I've had experince building solid state preamp and amp kits in the past. This is my first tube amp project (and I do admit I have bitten more than I can chew). The Super Reverb clone circuit was designed by Tracy Teague, a good friend of mine from Chicago, IL. He was a fairly experienced amateur tube amp builder. Unfortunately he passed away last December, before we had a chance to complete this amp. I am now trying to complete it on my own devices.

So let me recap ... based on the drawing I supplied, you are recommending that the current draw should read betweeen 56 and 64 mA. You are also suggesting that I monitor plate voltage as I do so. What is a safe plate voltage range in your opinion? I'll check it tonight and post the reading ...

Finally you are suggesting I up the value of the FB resistor ... I believe that the 820 ohm came from original Super Reverb schematic ... will try your suggestion and see what happens.

Maybe I should submit the whole wiring diagram and schematic for review by you amp building sharks. Feel free to pick on it as much as you like. I welcome any suggestions for improvement ...

Regards to all

Iggy

[Bruce / Mission Amps]

Yes, the plate voltage will vary as the idle current varies.
We don't know the actual B+ in any idle current condition and that makes this
conversation a little nebulous.
A good way to set proper power tube bias is to monitor the plate or cathode current, (as you are doing) and multiply that current times the actual measured plate voltage.
When measuring this via a cathode resistor, like yours, you are also measuring the current through the screens..
At idle, it is not very much in a 6L6 and you are erroring on the safe side anyhow.
A good starting point for a pair of 6L6s in push pull class AB is around 12 watts to 17 watts each.
Some like them colder and some like them hotter but I don't think I'd want them to idle over 19-20 watts each in any case.
You only have one current sensing resistor in your art work so the numbers you will see will be the total of the two power tubes.
In your example, with the two power tubes would be 24 watts to 34 watts.
Here is another method...


50% of tube DC rating (30 watts for a 6L6GC)

15 watts / B+ = idle current
EX:
15w / 460vdc = .033a or 33ma or, since you have only one current sensing resistor, 33mv times the two tubes... 66mv with your meter.

15 watts should be very safe for your amp.
I don't know your B+ so that's why I said 56ma to 64ma for the shared current through your current sensing resistor.

Your 820 ohm resistor, attached to an 8 ohm tap instead of the 2 ohm tap (which you don't have or are not using) is pushing too much NFB voltage to the PI and quite possibly killing the drive you are looking for.

[IggyRoyale]

Bruce,

I got around to rebiasing and taking some measurements. B+ you were enquiring about, measures 475V. Idle current draw that I am reading is 46mA. According to your equation my amp tubes are idling at about 475V x .046A = 21.85W. This is approximately 11W per tube ... in other words too low. I'll need to up the overal current draw a bit. The only problem is that the bias pot is turned up to the max and the ground DC voltage on the tube grid is reading only -47V. So ... either B+ is too high or I need to work out a way to get more -ve voltage out of my bias circuit ... am I correct?

Last question .... I am using 8 and 4 ohm taps on my output transformer. I have changed the NFB resistor to 3.3Kohm as you suggested (the only one I had within the range you specified ... thanks for the tip). In your last comments you hint that 820ohm resistor is more suitable to the 2ohm tap (which I don't have) This makes me think that the 4ohm tap should have it's own NFB resistor somewhere between 820 and 2700ohm .... is my thinking correct or am I being daft? If I am correct, can you please suggest a good value I should apply or at least how to calculate the right value?

Many thanks for your input guys! I have learned more in the past week than I have in the last year. Please keep your comments comming.

Regards

Iggy

[IggyRoyale]

Bruce and all,

Seeing that I haven't had any comments on this thread for a while, I am assuming that you all got bored of it ... never mind.

Just to close things off, I did get around to tweaking the bias circuit. Lowering a resitor value gave me enough control over the current draw to get the power tubes to operate at about 16W. The measurement across 1ohm resistor is now reading 68mA and plate voltage 478V. So, 478V * 0.068A = 32.5W Considering that there are two tubes in the power stage, this makes it 16.2W per tube.

This combined with changing the NFB resistor on 8ohm OT tap to 3.3 K and replacing a wrong resistor in the reverb (recovery) circuit has resulted in a rather dramatic improvement in overall volume and tone.

Thank you all for your contribution. This has been an amazing learning journey.

... speak in another thread ...

Iggy

[Enzo]

Some of us are not here on the weekends, I spend mine at home.