6G6 Bassman Output Transformer

[RightLurker]

Martin -

My thinking exactly. Your help and guidance are seriously appreciated.

Rex Herder
Bryn Athyn, PA

[RightLurker]

We're having quite the "snow day" here in the Philadelphia area, so I took some time to study the amp against schematics for the 6G6 and 6G6A. Martin had asked if the feedback resistor is 100K as shown on the 6G6 schematic. It is. I note, though, that the same resistor in the 6G6A schematic is 56K. Could this have anything to do with the fact that the 6G6 was intended to have only one 8 ohm speaker, whereas the 6G6A had two speakers wired for 4 ohms?

Martin also asked if the PI tail resistor is 47K as shown on the 6G6 schematic. I understand that this is the resistor soldered from lug 3 of the presence control to ground. The value of this resistor in the amp and on the schematics for both the 6G6 and 6G6A is 4700 ohms. I don't see a 47-anything resistor elsewhere in the amp or in either schematic. Is this the resistor you were referring to, Martin, or am I missing something?

[martin manning]

We're having quite the "snow day" here in the Philadelphia area, so I took some time to study the amp against schematics for the 6G6 and 6G6A. Martin had asked if the feedback resistor is 100K as shown on the 6G6 schematic. It is. I note, though, that the same resistor in the 6G6A schematic is 56K. Could this have anything to do with the fact that the 6G6 was intended to have only one 8 ohm speaker, whereas the 6G6A had two speakers wired for 4 ohms?

Martin also asked if the PI tail resistor is 47K as shown on the 6G6 schematic. I understand that this is the resistor soldered from lug 3 of the presence control to ground. The value of this resistor in the amp and on the schematics for both the 6G6 and 6G6A is 4700 ohms. I don't see a 47-anything resistor elsewhere in the amp or in either schematic. Is this the resistor you were referring to, Martin, or am I missing something?

Snowed a bunch here too. Kids are off school and I'm burning a couple of vacation days.

I think I wrote 4K7, which is another way to write 4.7K. I like it because when hand written there is no decimal point to be missed.

You are right on viz. feedback resistor; I missed that when I looked at the two schematics. The 6G6 and 6G6-A schematics show two different arrangements, with the 6G6 having a 125A5A OT and only one speaker. That's an 8-ohm transformer, and so it makes sense that the feedback resistor should be larger than the the one for the 4-ohm 6G6A to achieve a similar level of FB. The 100K (or 56K) and the 4K7 form a voltage divider, and the voltage on the 8-ohm system should be larger by ~1.4x. If you plug into an 8-ohm cab, your 100K will be about right, so try that and see how you like it. If you change to the correct OT for your 4-ohm cab, you should decrease that 100K resistor to 56K like the -A. Or maybe try 75K. This is something you could play with once you get the impedance match straightened out. The presence control can reduce the relative feedback of higher frequencies (it's a treble bleed for the FB signal), but the overall level of FB is controlled by the resistive divider.

MPM

[RightLurker]

Martin -

My mistake; obviously you wrote 4K7, and I should have seen it. Apologies. I ran the amp through an 8 ohm speaker, and it made little difference - none that I could hear, anyway. The presence control still does nothing. I think this amp has issues that I'm just not seeing due to lack of knowledge/experience. I'll keep futzing with it, though, and keep trying to learn more.

Thanks for everything.

RL

[martin manning]

The answer to the presence problem is there somewhere. I think you'll have to start looking at components... the presence pot, the 100K and 4K7 resistors, and the 0.1u caps to see if they are still functioning properly after all these years.

Solder joints? Maybe it's just not connected.

MPM

[RightLurker]

Back again. I've been looking at schematics and reading about how OTs work until I'm cross-eyed. One thing that jumps out is that Fender changed the values of the components in the feedback/presence circuit in many of the 6Gx amps sometime in '61-'62. For example, the 6G4 Super had a 5K-L pot, 10K load resistor, and a 1K5 tail resistor. In the 6G4A Super those values changed to 25K-L, 56K and 4K7. The 6G8 Twin used a 5K-L, 22K load resistor and 1K5 tail resistor; in the 6G8A Twin, those values changed to 25K-L, 56K and 4K7, respectively.

I jumpered a 56K resistor in parallel with the 100K load resistor in my amp (~36K net?), and it did slightly increase the effectiveness of the presence control. Any thoughts on decreasing the load resistor value further? Say to 22K or even 15K? Can I hurt anything by experimenting?

RL

[Structo]

No you won't hurt a thing by playing around with those values, unless it screeches and hurts your ears.

Even though the amps may appear similar, the Twin for example is a 4 x 6L6 amp, so the damping will be different on that vs a 2 x 6L6 amp.

It also depends which tap the feedback is taken from,

For example the Dumble ODS 100 w has the feedback taken from the 4 ohm tap with a 4K7 to a 2K pot with a 390R tail and 1uF cap to ground.

That setup works pretty good even on the 50 watt amp.

But I also imagine that the values on the PI also determine the values on the presence circuit.
It is a negative feedback circuit that can also be used for presence when a pot and cap are included adding or taking away some high end shimmer.

You can control the amount of distortion and tightness with the global feedback network.

Here is a great article about it.
http://www.aikenamps.com/GlobalNegativeFeedback.htm

[RightLurker]

Structo -

Thanks for the tip on the Aiken article. I'd already read it - let's make that looked at it - but thanks anyway. Most of it is over my head. Now all I need is a fist full of resistors!

RL

[martin manning]

Back again. I've been looking at schematics and reading about how OTs work until I'm cross-eyed. One thing that jumps out is that Fender changed the values of the components in the feedback/presence circuit in many of the 6Gx amps sometime in '61-'62. For example, the 6G4 Super had a 5K-L pot, 10K load resistor, and a 1K5 tail resistor. In the 6G4A Super those values changed to 25K-L, 56K and 4K7. The 6G8 Twin used a 5K-L, 22K load resistor and 1K5 tail resistor; in the 6G8A Twin, those values changed to 25K-L, 56K and 4K7, respectively.

I jumpered a 56K resistor in parallel with the 100K load resistor in my amp (~36K net?), and it did slightly increase the effectiveness of the presence control. Any thoughts on decreasing the load resistor value further? Say to 22K or even 15K? Can I hurt anything by experimenting?

RL

Those component values are not as different as they might seem. If we assume that all these 6L6 amps are running about the same plate voltage and are similarly matched to their speaker loads, they all have similar relative proportions for the NFB parts: The feedback resistor and the tail resistor are in a ~20:1 proportion for 8-ohm systems, and about 12:1 for 4-ohm systems in order that the feedback voltage is a similar proportion of the PI input signal voltage. In each case the presence pot is typically about five times the value of the tail resistor so that it doesn't significantly change the feedback voltage at the higher frequencies (as determined by the 0.1u cap) when it's turned down.

I can't help but think there is something amiss with the presence pot, or cap. We know the PI's tail is grounded, because you get sound, but maybe it's not a very good ground. Or, maybe the pot or the cap are not functioning correctly. You should be able to measure them in-circuit if you disconnect the NFB wire from the speaker jack. You can then check the operation of the pot, the FB and tail resistor values, make sure the cap is not shorted, and measure the resistance to ground from the pot housing (where the cap and tail resistor are soldered, and subsequently grounded through the pot's mounting).

MPM

[martin manning]

Oops! wrong button.

[RightLurker]

Martin -

The only thing I'm not certain about is the cap. The resistors are all within spec, the grounds are good, and the pot is working as it should.

I'm still in a quandry as to what to do about the OT. The 45548 is designed to run four 6L6 tubes into a four ohm load. A Bassman runs half the number of 6L6s (or 5881s), but is still driving a four ohm load. If you look at a 6G13 Vibrasonic, it's designed around a 45548 OT. It runs two 5881s into an 8 ohm load (single 15" JBL). This "works" because you're doubling the primary impedance, but also doubling the load on the secondary. (So suggests a book written by a well-known expert.)

Now for an example of what confounds me: Like all Fender amps, the Vibrasonic has an extension speaker jack. If you plugged another 8 ohm speaker speaker into the extension speaker jack, the amp would be running two 5881s through the 45548 OT into a 4 ohm load, which Fender apparently thought was just fine. So - is it really that bad that my amp has a 45548 running a 4 ohm load? I'm not sure I'm comfortable with the idea of pulling an original OT out of a vintage Fender amp.

[martin manning]

The only thing I'm not certain about is the cap. The resistors are all within spec, the grounds are good, and the pot is working as it should.

You know the cap is not shorted. Could it be open? You do get some effect from the presence, so maybe not. You'll have to lift one end to test it, or maybe temporarily parallel a new one over it with clip leads.

I'm still in a quandry as to what to do about the OT. The 45548 is designed to run four 6L6 tubes into a four ohm load. A Bassman runs half the number of 6L6s (or 5881s), but is still driving a four ohm load. If you look at a 6G13 Vibrasonic, it's designed around a 45548 OT. It runs two 5881s into an 8 ohm load (single 15" JBL). This "works" because you're doubling the primary impedance, but also doubling the load on the secondary. (So suggests a book written by a well-known expert.)

No problem there, except that its frequency response might not be optimum when used that way. When you plugged into an 8-ohm load you didn't hear a big difference, and the presence control still did next to nothing. That says it's not the mismatch that's causing the poor tone quality and presence to be ineffective, because in theory the 8-ohm load lined everything up properly. You reduced the FB resistor to 36K with the 4-ohm load attached, and got slightly more effect from the presence, but I gather it was not as much as you expect from a properly functioning circuit. That probably put the FB voltage closer to the right level for the 4-ohm load, but still it wasn't as good as you'd like.

You could test the OT to see if it is on spec; it's not hard to do. All you need is an AC source (even an AC wall wart power supply will do) and a multimeter. You'll have to disconnect the secondary to separate the windings, though. If the OT and the presence cap check out, then I'm running out of ideas.

Now for an example of what confounds me: Like all Fender amps, the Vibrasonic has an extension speaker jack. If you plugged another 8 ohm speaker speaker into the extension speaker jack, the amp would be running two 5881s through the 45548 OT into a 4 ohm load, which Fender apparently thought was just fine. So - is it really that bad that my amp has a 45548 running a 4 ohm load? I'm not sure I'm comfortable with the idea of pulling an original OT out of a vintage Fender amp.

Most amps can tolerate a 50% or 100% mismatch without damage, so Fender didn't make an issue of it. The tone quality might not be the best though.

MPM

[Firestorm]

I've seen a few of these so I'll weigh in: your amp was probably intended to be sold with the single-speaker 8-ohm cabinet (hence the Twin transformer, which would have been "correct" with the 2 X 5881/ 8-ohm load). Likely the music store (or someone else later) substituted the 2 X 12 cab.

I would not change the tranny (the Twin iron is much better anyway). Just replace the speakers with a pair of 16-ohm 12s, wired parallel (you don't want to risk killing the original speakers anyway).

As to why the amp doesn't sound right, check the tone caps and coupling caps. They're probably Astrons, they're probably leaking. I saw a transitional 6G6A where one of the .25 caps in the Bass Channel was passing 40 volts DC!!!

[dobbhill]

I'll second that, Firestorm. Leave the tranny, at least until you check for leaky caps. I have "blue" caps in my '63 6G6-B (4ohm output, SS rectifier).
D

[RightLurker]

Good advice, firestorm and dobhill. After a good deal of thought, I've already decided to leave the transformer in the amp. I'm going to be experimenting with some different component values in the NFB/Presence circuit, and I'll check the tone and coupling caps - which are, indeed, Astrons. Eventually I'll replace the speakers with a pair of 16 ohm units; for now, I've disconnected one of the two speakers, giving me an 8 ohm load. I don't turn the amp up to more than 3 or 3-and-a-half, so there is little chance I'll damage the speaker. Thanks!

RL