GA5 Les Paul Jr - volume question

[yorgle]

A few months back, I built a ga5 LP Jr (6sj7 preamp version) using the schematic in the library.  As far as tone, it sounds absolutely beautiful, but it just seems a bit quiet compared to other low-watt SE amps I've played.  It also stays crystal clean all the way to 10 unless I'm really doing some hard strummin'.  The amp is mounted in a small cab (actually an old typewriter case) with a 10" 8ohm speaker from an Electra amp.    Perusing the internets, I hear lots of folks say things like "much louder than I expected" referring to original Gibson ga5's and so now I'm wondering if something is amiss with my build.  I certainly do not expect this amp to rattle the windows, but as it sits, it doesn't even rattle the dog sleeping 2 feet away. 

For point of reference, I plugged the same speaker into one of my other SE amps and it's plenty loud.  I also tried changing out the cathode bypass cap on the 6v6 from 20uf to 100uf without any noticeable effect.  While I was in there, I took some measurements:  B+ is 360; v1 (6sj7) has 150 on the plate, 20 on the screen and cathode is grounded; v2 (JJ 6v6GT) has 353 on the plate, 273 on the screen and 16.7 volt drop across the cathode resistor (470R).  By my maths, that's over 12.5 watt, so kinda on the hot side, which would explain the tall headroom.  Would that also explain the lack of volume?

[yorgle]

This is the schematic I followed:  https://el34world.com/charts/Schematics/files/Gibson/Gibson_les_paul_jr.pdf  The only changes I made were eliminating one of the inputs and adding an extra triode (6sq7) for a tremolo. 

[shooter]

does a boost/gain pedal get you there?
I never finished my pentode preamp class so I can't help there 

[nandrewjackson]

I was gonna suggest the inefficient old speaker as the problem,  but you've covered that.


I'm not familiar with that 6SJ7, surely someone here is and will be with you shortly. 

[66Strat]

I think that the 6SJ7 grid leak bias may be holding the amp back. Operating points from the RC19 Tube Manual show Voltage Gain up to 263 Vo/Vi is attainable with cathode bias operation. See attached.

[PRR]

> 6SJ7 grid leak bias may be holding the amp back. ......Voltage Gain up to 263 Vo/Vi is attainable with cathode bias

Gridleak bias "can" give as much gain as cathode bias. His voltages look ballpark to me.

If the G2 bypass cap is weak or missing, gain will be quite low.

The max gain is actually limited by load. 6V6 grid resistor. Should be 500k Max. Here it will be 1Meg when full-up, oops. Oh well. Taking 300V supply, 220k Rp, 1.0Meg Rg, gain can be 98. Any way you count, input sensitivity will be 50mV-100mV, compared to 20mV on classic rock amps. You gotta strum 3 times harder to slam the 6V6. Is what it is.

[yorgle]

PRR:  I'll try swapping the G2 capacitor on V1 tonight and see if that helps.  I'm not sure what you mean by "Here it will be 1.0meg when full up, oops.  Oh well."  Could you please explain that statement?  Thanks.

EDIT: ok, I think you're referring to the 1meg volume pot, and saying that it should be 500k max, right?

[bmccowan]

This is the schematic I followed:  https://el34world.com/charts/Schematics/files/Gibson/Gibson_les_paul_jr.pdf  The only changes I made were eliminating one of the inputs and adding an extra triode (6sq7) for a tremolo.

Do you have a schematic or sketch of what you actually built. If you build an amp to the schematic you posted, it'll be plenty loud enough to wake that dog. Its the same circuit as the 5C1 champ. But adding your tremolo 6sq7 circuit may have produced unintended consequences.
And Joe Piazza always drew nice clean schematics, eh?

[yorgle]

This is the schematic I followed:  https://el34world.com/charts/Schematics/files/Gibson/Gibson_les_paul_jr.pdf  The only changes I made were eliminating one of the inputs and adding an extra triode (6sq7) for a tremolo.

And Joe Piazza always drew nice clean schematics, eh?

Here's the schematic showing the attachment points for the 6SQ7 tremo-o-nator.  The red circle is the optocoupler.  The resistor is the plate resistor for the trem tube.

[bmccowan]

Cool that you kept it all octal. Not enough in that sketch to see if there is a potential leak to ground, but have tried disconnecting the trem circuit to see if you get expected volume? And I assume you have tried swapping in other tubes?

[yorgle]

I've swapped power tubes but don't yet have an extra 6sj7 (ordered one yesterday).  I'll draw out my tremolo circuit and post that later today, if possible. 

[ChopSauce]

Just a side note: I've read similar complains about the lack of headroom about the "vintage" Vox AC4 amp from the sixties, which shares the exact same topology : a pentode for pre-amp straight into the power stage (SE/EL84).

Many users don't complain but Doug from https://el34world.com/Forum/index.php?topic=21346.0 also found his amp too quiet... 
(hence the linked revision)

I'm not sure that helps but I'm sure I won't trust everything from all these guys claiming their build is awesome (eg: very loud) all over the internet... 

[bmccowan]

Yorgle is not complaining about a lack of headroom, but the opposite.

It also stays crystal clean all the way to 10 unless I'm really doing some hard strummin'.

Gibson made several models with the 6SJ7 in v1 position, and none of them stay crystal clean past about 2. And "louder than expected" is different than "very loud." So I am all for mods and tweaks to increase/decrease just about anything - but not until I know that the stock circuit is working correctly and that the tubes are good. You see the "not very loud" comment often on the forum for all kinds of amps, and the answer is almost always a mistake in the circuit like a misread resistor value. I did it just recently when my tired eyes put in a 150 ohm resistor to ground instead of the intended 1.5K. Damn, that amp was not as loud as I expected.

[PRR]

Here's the schematic showing the attachment points for the 6SQ7 tremo-o-nator.  The red circle is the optocoupler.  The resistor is the plate resistor for the trem tube.

No, you left out important details. If the red circle LDR gets ANY light, ambient or from LFO bleed, it hurts volume. (Disconnect LDR to check.) If you are saying that trimmer is the LFO plate resistor, I'm baffled.

What did you really build?

[yorgle]

PRR:  sorry if my drawing is confusing.  The trimmer is the trem's intensity pot.  The plate resistor is only there to show what I modified from the stock schematic.  I'll try pulling the LDR off tonight and see what happens. 

[bmccowan]

I think what both PRR and I are trying to say is that your sketch is incomplete. If I consider the resister you added to be point A, and the pot to be point B. There has to be a bunch of stuff going on between those two points. Completing the sketch would help. BTW, this is a nice video of how that vintage GA-5 LP Junior sounds. https://www.youtube.com/watch?v=B03vhEqZa8Y
Cheers,

[yorgle]

You guys were right- the dog is now awake!!  I temporarily disconnected the LDR from the volume pot and WOW- instantly louder.  So the problem is definitely that the tremolo was shunting too much signal to ground. 

[bmccowan]

Great! Yay! So assuming you still want tremolo, if you sketch out the rest of the circuit, we may be able to help you identify the problem. Regardless that simple circuit is a good one, in my opinion, so you should have a cool amp. Love the typewriter case idea! I went off to the University of Maine in 1971 with a few changes of clothes and a Smith Corona.

[yorgle]

I found the problem with the tremolo- I realized while I was drawing out the circuit that my on/off switch was wired incorrectly.  The way I had it lifted the ground connection to the speed pot to turn the trem off, which worked great to stop the oscillations, but the LED remained steadily lit, causing signal to always be grounded regardless of whether the trem was on or off.  So I moved the switch to lift the ground on the photoresistor instead so now when the trem is off, the signal stays at 100%.  With the trem on, I lose about 25% volume, but I believe that’s unavoidable.  Here’s my revised trem circuit.  Note that I temporarily replaced the intensity pot with a fixed resistor to simplify things while testing. 

[yorgle]

Being able to open the back of the case is very convenient.

EDIT:  the trem tube in my drawing above is actually a 6sq7, not a 12.  Oops. 

[bmccowan]

Very cool - love the case. Damn I donated my Smith Corona with its case decades ago. I really should learn to hoard everything!
I trust you now like the amp with the tremolo off?
I am not sure that you need to lose 25% of volume with the trem on, and I'm not sure about your switch placement. PRR will know better than me. In the meantime - did you reference this site? http://www.valvewizard.co.uk/trem1.html

[yorgle]

I messed around with the trem some more tonight and settled on a 7.5k resistor in series with a 10k intensity pot which yields a happy balance between depth and loudness.  The dog sleeps further from the amp now!