Ampeg "Guitaramp", 1956?

[jjmt]

jjmt,

for each tube, following the process explained by Uncle Doug in his youtube video:

- (power off) measure the resistance between pin 8 of the 5Y3 and pin 3 of the 6V6. This is the impedance on half of the input side of the output transformer ®
- (power on) measure the voltage drop across that same region (V).
- calculate plate current as V/R (PC)
- measure plate voltage relative to ground (PV)
- calculate Power Dissipation (PD) PD = PC*PV = 10 watts

6V6 tube has a max rating of 12 watts. 10/12 = .83

I installed independent bias resistors (550ohms and 680 ohms) and capacitors for each 6V6 to make the outputs approximately equal at 10 Watts. The amp has plenty of volume, so you may be correct that I can reduce the output of the tubes further.

I did not chart the results after I replaced the cathode bias resistors and caps, I just assumed that the results would fall out of my plot at 10 watts per tube (red dots on plot I posted earlier). To get down to 70%, I would be at 8.4 watts. To do that I think I would need 680 and 1000 ohm bias resistors by interpolation for the outer tube and extrapolation for the inner tube.

I am happy to hear your thoughts. Am I on the right track in your mind?

Larry

Hi. That all seems correct, although the splitting of the cathode resistors might have been a bit of overkill. You haven't mentioned what your plate voltages are, but some quick guestimates put the difference in plate current of your two tube at under 5ma originally. That's on the high end of what you can get away with, but dropping the joint current draw by 30% would likely have gotten you down to around 3.5ma, which is pretty standard for a "matched pair", at least in the current state of NOS.

In any event, if your tubes aren't red-plating, then I would leave it at the current levels. On a different subject, you didn't mention if you by-passed the split resistors? The original circuit had a bypass cap to eliminate interaction of the two tubes that can result in some nasty oscillation. If you bypassed the independent resistors, all is well, and if not, all is also well . But without the bypass caps your plate impedance is going to increase (I think), which is going to affect what it sounds like. More compression, I think. Maybe that is a good thing, but just curious as to your experience.

Beyond that, there is the whole Class A versus AB discussion that is a minefield that I usually try to avoid, but there are those that argue that a shared cathode resistor effectively imposes class A on the amp (much debate can be found on this), whereas splitting the resistors is more consistent with class AB, which would support a bias level down towards 70% and which would allow your tubes to run cooler. Given that you seem willing to experiment, you have somethings to consider.

[vibrating string]

Plate voltages are measured at 346v on both tubes. I replaced a single 15Watt? wound cathode bias resistor and cap to ground with a 10Watt resistor and cap on EACH 6V6 cathode to ground. I forget the cap value (I dont have my notes with me), but I duplicated what was there. My workbench has a dismantled guitar on it at the moment, but I will get back to the Ampeg to plot my power dissipation and plate currents when that is off the bench....waiting for parts.

I get lots of sound at volume setting 2-3, so maybe increasing the bias resistors would not hurt. Also, I am pretty certain that the speaker is original. I might shop around for a nice recone kit or a replacement.

Larry

[jjmt]

I get lots of sound at volume setting 2-3, so maybe increasing the bias resistors would not hurt

Fiddling with the power tube bias is not going to change this. The bias just determines when the power tubes will distort, and what it sounds like when they do. What you describe is just a volume pot thing. Maybe it has a linear pot and you could replace it with an audio taper?

Also, I am pretty certain that the speaker is original. I might shop around for a nice recone kit or a replacement.

I'm kinda impressed that you would consider doing this yourself. If you proceed, please describe your experience?

[vibrating string]

Volume pot....OK. I will look further into this. I am not certain, but I recall testing the pot that was originally in the amp and it was linear (500k at position 5). You have me thinking to go back and recheck things.

I have re-coned two Advent speakers in the past and replace the apron on another speaker. The tolerances are tight on the coil/magnet and I was very aware of the delicate nature of the process. There are good YouTube vids on the process.

Larry

[Hangman]

Its an interesting exercise, but why tailor your amp to your tubes? I mean if you put a different pair of 6V6s in there, they could very well perform differently, are you going to change the cathode resistor every time you change the tubes?

[vibrating string]

That is a good point. I thought about that too. I am still learning. I have never changed a tube in either that amp or in the Gemini. I just don't use them that heavily. But I am thinking I will go back to a single bias resistor and bypass cap. The power dissipation calculation for the tubes differed by about 15% so I thought I would even that out somewhat. I am not sure it made a difference.....sounded good either way (one vs. two bias resistors values). I will consider going back to one bias resistor and cap. Using a single bias resistor and getting matched tubes would be a good path when new tubes are required. I might leave it as is. I left a note in the amp.

I am just finishing upgrading a SilverTone strat guitar for a friend. The replacement parts came over the past few days, so I can move on to other issues. Also, I found that the treble pot on the Gemini amp is breaking up in the high treble where the pot hits the switch to high treble. Some contact restorer might fix that, but I have to open the thing up to get to it.

Then there is redoing the basement bathroom, but that is another story.

Larry

[jjmt]

Its an interesting exercise, but why tailor your amp to your tubes? I mean if you put a different pair of 6V6s in there, they could very well perform differently, are you going to change the cathode resistor every time you change the tubes?

If it's an amp for home use and isn't going to get that much playing time (like a lot of really old amps), the power tubes will last a good long while. If you've got a pair of RCA black plates (or whatever) that you really want to use, then why not? I doubt I would do it, but then again, I can usually find a matched pair in a box somewhere...

[vibrating string]

So the Ampeg is back on the bench. I decided to do a couple of things:

- change back to one cathode bias resistor and cap shared between the two 6V6 tubes. The value was 300 ohms originally, and I changed that to 680 ohms to lower the plate current to get the tubes output power down a little (explained earlier).
- replace the speaker with a Ted Weber replacement for an original Jensen P12Q. I am waiting for the speaker to arrive.

I decided against rebuilding the old Jensen speaker, which actually sounds quite excellent, but I thought I would never know if the speaker was holding me back unless I try this swap.

Larry

[vibrating string]

Slider,

I calculated the 6V6 plate currents using the process I described in a previous entry to this thread. It is based on the plate voltages and the impedance of each half of the input side of the output transformer for each 6V6. The plate currents are in the mid 20ma range for both tubes using a single 680 ohm bias resistor (22 and 26 ma). (I actually measured plate currents for 470 and 770 ohm resistors, so I am interpolating for 680 ohms.)

The original speaker is labeled P12Q on the frame. I passed this information to Weber Speakers for the replacement and they spec'd out what they say is a good Weber alternative. I have to have faith.

Larry

[vibrating string]

Thanks Slider....I wish I chatted with you before I purchased the speaker. CJ gave me three choices. Here are the descriptions:

12A125 (8 ohm, 30w, light dope) this is the main P12Q variant from the mid 60’s, with a multi-ribbed seamed cone for very emphasized treble and headroom. This is the standard 12A125 which works great in tweed and blackface style amps.

12A125A (8 ohm, 30w, light dope). 11oz alnico magnet / 1.25” voice coil. Authentic sounding choice for woody, reedy, late 50’s / early 60's Fender tone. Tight low end, detailed high end. Compressed and fattened at higher volumes.

12A125S (8 ohm, 30w, light dope) same 11oz alnico magnet and 1.25” voice coil but with a different style mid 60's cone which is more forgiving than the standard one. moderate breakup, good moderate tone, very versatile.

Larry