Express Harmonics and Feedback

[gearhead]

Decided to start a new thread, now that I have my Express up and running.

Spent a good deal of time on it this weekend, fancy that!

Better tubes do make a -Huge- difference. Have a mix of Brimar and NOS JAN Phillips 12AX7s in there now, along with a set of EL-34 Tesla's. Gonna pop in the Xf3 Mullards soon. Still dialing in the tone/sound; haven't gotten to changing any components because wanted to try everthing else first.

One thing that's got me a bit puzzled is that mine seems a bit too much under control. It definitely roars and would never think this about any other amp than a trainwreck. I can get feedback, but have to wrestle the neck a bit to get it. Was expecting to have to quiet the beast down, but that's not the case so far.

I've chopsticked some wires around to do the exact opposite of what most do; get more interaction.

Is there anyway to make it LESS stable? LOL. Reduce effects of NFB?

Thanks all!

[Lonely Raven]

I asked almost the same thing about my Ceriaton Express. I Chopsticked around the pots (as suggested) and actually got different feedback results!

Apparently the circuit is very interactive!

I don't have much (almost none) time to play the amp at the volumes it was built to play at, so I've not completely dialed it in yet. But I was surprised how much interaction there is between wires, tubes, cabs, and speakers.

[gearhead]

Do you have a link to that discussion? I tried a number of different searches on feedback, etc, but never hit the right combination other than Jackie Treehorns Liverpool observations (flying treble lead).

BTW, I did try a bit and think it made a difference! Will do more, but any other area?

I also put another couple hours on the amp, and it is definitely getting tonally sweeter!

[gearhead]

Heh Guys,

Been poking and prodding, chopsticking a-plenty, and other than some slight expansion of feedback and harmonics around the tonestack pots, no big change.

Is there a possibility that somewhere it's leaking to ground, but not causing any tell-tale voltage drops?

I was expecting to be riding a cross between a bucking bronco and a wild-eyed thoroghbred, but such is not the case.

My Express is TOO well behaved.

Any ideas?

[skyboltone]

Heh Guys,

Been poking and prodding, chopsticking a-plenty, and other than some slight expansion of feedback and harmonics around the tonestack pots, no big change.

Is there a possibility that somewhere it's leaking to ground, but not causing any tell-tale voltage drops?

I was expecting to be riding a cross between a bucking bronco and a wild-eyed thoroghbred, but such is not the case.

My Express is TOO well behaved.

Any ideas?

I know what you mean. In the two hole liverpool when I first fired it up it was that bucking bronco you describe. As I tweaked and eliminated the noise and hum I noticed that it did become a little calmer and more sedate. I think the first thing I would try is V1. Tungsols are known for high gain, why don't you try one of those? EH too. What's the voltage at B+4? I think it's a matter of gain really. Have you compared your layout to Francesca? As a matter of fact, post the whole voltage magilla, plates and cathodes all the way through.

[gearhead]

Here they are. Did put in a Tung Sol in V1 before these measurements. Happen to have a VTV tester, and all the preamp tubes I put in have a gain of at least 95, and most 100+.

Odd things either new, or didn't notice before. The B+1 changed when engaging the rest of the circuit (on). Measuring the PI grids, momentarily, eliminated ALL hum. All controls at the 1200 O'clock postion during measurements.

Power In: 122 VAC
PT Secondary: 614 VAC

Bias - 30.5 VDC

Heaters: 6.3 VAC Tap to Tap, 3.0 VAC Tap to Ground

B+1 426 VDC on standby -then- 412 VDC when switched to on
B+2 399
B+3 317
B+4 297
B+5 278

V1a plate 195 VDC
V1a cath 1.94
(tonestack driver)


v1b plate 170
vab cath 1.52
(first input from jack)

v2a plate 249
v2a cath 2.64

v3a plate 205
v3a grid 12.4 (took away all hum when measured)

v3b plate 197
v3b grid 10.7 (took away all hum when measured)

V3 common cath 34.1

V4 plate 408
V4gs 387
V4 grid in -25.5

V5 plate 408
V5gs 386
V5 grid in -25.5

Thanks!

[Phil_S]

When you close the standby switch, the downstream part of the circuit represents a significant load. It is normal for B+ to drop and 412v is your operating number for B+1.

I am seeing mostly similar preamp voltages on my Express, but not getting comments on my post. The big difference between yours and mine is on V3 (the PI). In round numbers, I'm seeing 12V on the cathodes and 8+V on the grids. Did you build this to the Francesca schematic? Where do you have the Presence set when measuring?

[gearhead]

Not sure if I measured the B+1 under load before. Thought I had, and don't remember such a difference, but memories can lie. Didn't write it down.

Built to Francesca specs, and almost all the layout is the same; or darn similar. I do have some grid lines that run -shorter-, but that was me trying to eliminate noise from the get-go.

PI voltages have me baffled. Tried to tackle understanding the precise electrical cause and effect of shared cathodes as 180 out of phase, but just couldn't fit it all in my head at once. LOL.

Presence set at 12 noon, like all of them.

Another thing: my power filter cap discharge rate is different than it used to be. For safety sake, would check it and wait till they were discharged to start working on the amp. Used to take about a second for 5-7 volts, starting at 400+ and discharging to near zero at the exact same pace per second. Now, the voltage on B+1 caps drop from 400+ to 300 or high 200s in about a second. After that, it does the 5-7 V per second drain. Maybe they are breaking in?

Thanks!

[skyboltone]

That the hum goes away when you measure grid voltage is no surprise. If you used a VTVM or a scope you'ld find less impact. If you had a signal going at the time of measurement I think you'ld find that the gain all but goes away when measureing grid voltage. The 25 volt difference between V1A and B makes sense when you look at the difference in cathode resistors. My liverpool they were about the same.

Seems I read on the forum here that occasionally even KF would have trouble with an overly tame amp. Try to get the V1B plate next to the V1A cathode. Should raise some kind of stink! Oh, and take the shielded cable off the volume pot just for grins if you have one there at all. You could try to remove it from the input too.

I don't remember exactly but aren't the PI voltages a little low? Are you sure that's a 9.1K? in the dropping string? Just thinking aloud.

[gearhead]

Thanks SB!

The B+3 is the result of two 9k1 resistors, which is 317V. The PI plate voltages are subject to asymetrecial plate resistors of 82k and 100k. Don't quite know if they are low or not?

AFA tame, maybe I should just rip out each wire one by one, and replace with long, parallel, flying in the air wires. LOL.

[Firestorm]

I think it's interesting that KF would up with exactly the PI component values used in late tweed Fenders (and thus in early Marshalls). Very much an "if it ain't broke, don't fix it" kind of thing. But this might be a fun (and pretty simple) area to experiment with. In some blonde amps, Fender got the tail resistor down as low as 6K8 (with a 12AX7), before turing tail (heh) and going up to 22K and even 27K (with 12AT7s).
Randall Aiken suggests replacing the tail resistor and the cathode resistor with trim pots and playing around with them to get the desired effect. Lower tail resistor values should throw off the (already tenuous) balance of the two sides a little. Large values will make them more balanced, but eventually reduce headroom. One of the things I'll probably end up fooling round with.

[jjman]

I think on TGP I read how someone achieved much better sustain/feedback from flipping the speaker polarity. (Not merely the OPT secondary since that would turn the negative feedback into positive.) Its the “absolute phase” thing. He swears it was a huge difference.

Probably not a TW btw.

[Richie]

I think it's interesting that KF would up with exactly the PI component values used in late tweed Fenders (and thus in early Marshalls). Very much an "if it ain't broke, don't fix it" kind of thing. But this might be a fun (and pretty simple) area to experiment with. In some blonde amps, Fender got the tail resistor down as low as 6K8 (with a 12AX7), before turing tail (heh) and going up to 22K and even 27K (with 12AT7s).
Randall Aiken suggests replacing the tail resistor and the cathode resistor with trim pots and playing around with them to get the desired effect. Lower tail resistor values should throw off the (already tenuous) balance of the two sides a little. Large values will make them more balanced, but eventually reduce headroom. One of the things I'll probably end up fooling round with.

I had a very early fender [untouched]5f6 bassman here,and it had the 6.8k tail.
And it also used different tone caps values, it didn't use the .022..it was more similar to later fenders

[geetarpicker]

On a different note, what kind of cabinet/speakers are you running?

My original Express and two clones all behave quite differently depending on what cabinet/speakers I use, all other things equal. Some speakers have nice controllable feedback all around, and others are so uneven in the high end that the feedback likes to center around one note no matter what note you are trying to get sustain on. This is one other reason I favor the old pulsonic cone G12M (or Scumback clone) as the sustain is more alive. Another thing I've noticed is the better response I get by taking the amp off the cabinet. If the amp is stacked sometimes moving closer to the speaker to encourage feedback is negated by hum pickup from the head which chokes the notes. If I am trying to get feedback at lower volume levels (like attenuated 8-12db or something) I'll move in really close to the cabinet AND move the head away so the hum doesn't increase when I get in close.

[gearhead]

So far, have run it thru two closed-back cabs; a George Scholz (Joyce) 2x12 cab with GS speakers, and a Jenkins 4x12 with Eminence Governors. Know what you mean about certain notes; when I do get feedback they have different feedback ranges between the two cabs. I do have an open back 2x12 with Scumback H75s, but was shying away from open backs. Maybe I should give that a go!

WRT internal changes, just for kicks, was thinking on running the V2 grid wire parallel to one of the V1 grid wires. It'd have to be the jack input grid wire (V1b), because isn't the V1a out of phase with V2 (and defeat the purpose)?

Did try moving the NFB from 8 ohm tap to the 4 and 16 (easier than swapping out the 100k resistor). Very slight diff.