Is my OT fried?

[Tom Phillips]

...To my eye this looks most like a plexi TS, but with tweaked values. And actually while waiting for my 500k pots to arrive, I actually have more typical plexi 250k/1M/25k pots...

Look at that 560k resistor that you show across the treble pot in your schematic. That is going to drastically affect the pot taper. (Check out RGs secret life of Pots article to understand the effect) In reality I think it is just a mistake as shown.

If you use Duncan's tone stack calculator you will see a similar resistor wired from the treble pot wiper to ground. I believe it is included in the tone stack calculator to represent the load on the tone stack.

[Gainzilla]

Thanks Tom. I may remove the 560k from the treble pot for now, at least until I get it working properly. I have read the secret life of pots, but it's been a while. Perfect time for a refresh.

Here is a screenshot from Duncan's TSC (running on a Mac!).

• Green is Standard Marshall
• Yellow is what I've got in there now (250k/1M/25k)
• Blue is as per the schem.

It's weird. As currently wired, the treble pot goes from anemic to loud and midrangey. The Mid pot goes from heavily scooped, to nasal. Bass pot is highly ineffective. all the way down is a nice crunch (Think AC15 with a Tele and you'll be in the ballpark.). All the way up and it's tubby and like the highs have just been rolled off. Something is definitely not right.

I've noticed on some schems, the "not connected" pin of the bass pot is linked to pin 2. Mine is currently not connected.

[Tom Phillips]

I don't think the difference between the green, yellow and blue response plots is significant enough to cause the symptoms that you have described. At lease not for the mid pot rotation settings that are shown. However, your description of the wiring, the schematic in post #57 and the tone stack schematic shown in post #62 are all different. I expect that the tone stack will work fine when you get that sorted out. One good approach would be to first correct the working copy of your schematic and then trace out the physical build and mark off the schematic with a highlighter as you go.

[Enzo]

Connecting the third, unused leg of the bass pot to the center leg - the wiper - makes no difference at all. The pot is used as a variable resistor.

[Gainzilla]

Thanks for the info guys. I'll play with it tonight and see if I can get it working right.

Cheers!

[Gainzilla]

deleting duplicate post

[Gainzilla]

Hi all,

I looked over the tone stack wiring and made sure everything was good, as per the schem. Everything looks fine. The more I played with it, it sounded like the tone stack was interacting with something in a weird way. If you look at the schem I'm following, there are two mosfet stages, the cathode follower driving the tone stack, and the fx loop send directly following the tone stack... Any chance one of those could be contributing to the issue?

I bypassed the stage and went directly from the tone stack to the send jack, and got wild oscillation squealing. I'm assuming at that point if I flipped the OT primaries that would go away, but not sure. While I was there I was able to turn the master down enough to hear the tone without squealing and I was able to hear the tone stack working withing what seemed to be normal parameters. Also, there is something about the dual ganged pot that is supposed to keep the signal from getting out of phase at the beginning. When I bypass the fx send stage and flip the gain stage switch to go directly into the low stage, it oscillates. I think it made other weird sounds when in high gain mode and the gain turned up. Not squealing, but a sort of squishy mushy low volume fuzz (if I remember correctly).

So, not sure if it's just a normal thing that I encountered when bypassing an entire stage without respect to phase, or an indication that something else is going on.

Thoughts? Do I even need the fx send stage? I think some amps get by without (Budda superdrive series come to mind).

Cheers!

btw, when I'm able to get the tone stack dialed in there is a pretty magical tone happening. Just wish I could dial in a little more top end.

[Tom Phillips]

...I looked over the tone stack wiring and made sure everything was good, as per the schem. Everything looks fine...

Do you still have that 560k resistor connected as shown? I'm not sure that you got what I was trying to say in my post #61 so I say again a different way...that circuit configuration is wrong unless your intent is to drastically alter way the 500k treble pot performs. This could be most of or all of your tone stack weirdness. Compare your schematic to the one shown in Duncan's TS calculator.

Try a test with the 560k resistor out of the circuit. If things seem unstable then put a 1Meg resistor from the treble pot wiper to ground. (Not across the legs of the pot)

[Gainzilla]

Whoops, nope I got what you meant. Think the communication gap was all mine. When I said I wired it up as per the schem, it was regarding wiring up the pots. However, I did simplify by removing the 560k. I did install a 560k from treble wiper to ground, which I viewed as more of the reference used to bias the next gain stage (Ie, not necessarily seeing it as part of the tone stack). It's still unstable with a 560k. Also, it's a Mosfet stage so really have no idea if the ground reference even matters. I'll bump it up to 1 Meg and see if that helps.

Thanks!

[Tom Phillips]

…I bypassed the stage and went directly from the tone stack to the send jack, and got wild oscillation squealing. I'm assuming at that point if I flipped the OT primaries that would go away, but not sure…

The stage you bypassed is not within the power amp’s negative feedback loop. Therefore, that change will not affect the phase of the feedback signal and there is no need to swap the OT primaries due to the bypass. Are you absolutely sure that the feedback phase is negative?

…While I was there I was able to turn the master down enough to hear the tone without squealing and I was able to hear the tone stack working within what seemed to be normal parameters…

Seems like a wiring dress issue to me. Not uncommon for high gain builds. Got build photos to post?

…Also, there is something about the dual ganged pot that is supposed to keep the signal from getting out of phase at the beginning…

No. The dual pot will not affect the phase of anything. However, a wiring dress issue to be concerned about is that fact that, in order to connect to the dual pot, you are routing signals from output of one gain stage close to the output signal of a subsequent gain stage. Again this is a wiring dress issue requiring careful layout.

Your schematic does not indicate which power supply nodes power what. Are you by chance powering all the preamp nodes and FETs from the last power supply filter node? That could cause stability problems because the various nodes would not be adequately decoupled from each other. That would allow the signal to ride on the power supply rail between stages. The fact that you reduced the values of the series resistors between filter nodes also reduces the decoupling between power supply stages. Usually people just talk about “filtering” but “decoupling” is also important. The stages of classic amps do share the same power supply node but the sharing is not universal.

…Do I even need the fx send stage? I think some amps get by without (Budda superdrive series come to mind)...

I think you do if you want a good FX drive stage. I think a more important question is do you need the FX stage at all. You could disconnect it and wire around the whole stage for now and work on getting the amp stable without it first. Don’t just jumper around it. Power it down and disconnect both the in & out connections.

Those are my thoughts anyway.

Hope they help,
Tom

[Gainzilla]

Are you absolutely sure that the feedback phase is negative?

No. Not really sure how to check that, but I can take a guess. I would assume that the phase remains consistent until you do something like take a signal from a cathode (cathode follower stage flips phase). If that's true, my signal probably started off positive for the first 2 stages, then flipped negative and remained that way until the phase inverter, which flipped it positive again? Am I even close? I'll do a little googling and see if I can find out.

Seems like a wiring dress issue to me. Not uncommon for high gain builds. Got build photos to post?

Sure. Disclaimer 1... The input jack is definitely in the wrong spot, but I need to fill the giant gaping hole left by the fender pilot light assembly before I can move it... Should be this weekend (JB weld a thick washer, then install a smaller LED pilot light where the input jack is.

That said, I'm using shielded cable to go from input jack to the switch, then from the switch to both the low and high input stages. The shields are twisted together where they meet at the switch and attached to the bus bar.

Disclaimer 2... This is probably way too complicated a build to squeeze into such a small chassis, compounded by the fact that the existing holes somewhat limited what I could do.


Note: Looking at the amp from the front, from Right to left the controls are Gain, Treble, Mid, Bass, Master. Presence on back next to output jack.

Your schematic does not indicate which power supply nodes power what. Are you by chance powering all the preamp nodes and FETs from the last power supply filter node?

My B+ nodes are arranged like so:
B1: Output Transformer
B2: Power Tubes
B3: PI and Loop
B4: Preamp

Don’t just jumper around it. Power it down and disconnect both the in & out connections.

Thanks Tom, excellent advise. Will do.

[Tom Phillips]

In response to “Are you absolutely sure that the feedback phase is negative?”

No. Not really sure how to check that, but I can take a guess. I would assume that the phase remains consistent until you do something like take a signal from a cathode (cathode follower stage flips phase). If that's true, my signal probably started off positive for the first 2 stages, then flipped negative and remained that way until the phase inverter, which flipped it positive again? Am I even close? I'll do a little googling and see if I can find out.

The short answer is “No. You are not even close” But you will continue to learn.
Each standard triode amp stage causes a ~180º phase shift and a “follower” stage has ~0º phase shift. Your googling on the subject should turn up sites that explain this with nice illustrations that will help clarify what’s going on.

The feedback I’m talking about is the negative feedback loop from the speaker output jack back to the phase inverter. See the attachment of your schematic with a blue line I added showing the negative feedback connection.

There is an easy test you can do to determine if the feedback phase is correct. You want negative feedback. If the feedback is inadvertently positive then the amp will be, at best, unstable and, at worst, produce a loud howl.

With the amp powered off disconnect the feedback loop. An easy way to do this is to remove the loop connection at the speaker jack. Leave the end of the wire handy or connect a clip lead to it for use in the following steps. Just make sure it doesn’t contact anything else in the amp. The rest of this is done with the amp powered on so only proceed if you know you can do it safely. Connect the amp to a speaker, feed a signal into the amp and adjust so you get sound from the speaker. You can use a signal generator, guitar a CD player or any other signal source. You can even just lay the guitar down so it picks up noise such that you hear a buzz in the speaker. The volume can be at any comfortable listening level. Now connect the loose end of the feedback wire to the jack where it was originally connected. The volume of the amp’s output will either go up or down when you connect the wire. If it goes down the feedback is negative and all is well with that part of your circuit. You have verified that the OT connection is phased correctly. If the volume goes up when you reconnect the resistor then the feedback is positive and you need to reverse the OT leads to change to negative feedback. Note that, if the feedback is positive the amp may squeal or howl when you connect the resistor which is another indication of positive feedback.

[Enzo]

To add to Tom's explanation:

Unless it is a cathode follower, each stage is a phase inversion or polarity reversal if you like. So if the input is positive, then the signal leaving will be negative, and the next stage will re-invert it to positive again. So even numbers of stages will keep the positive phase. And then odd numbers will invert it.

This normally is not an issue until you combine things. Some people like to plug into two channels at once on an amp. In some fender amps, this is OK, like when both channels have the same number of tube stages. But in some other models one of the channels has two stages and the other three. SO if you plug the guitar into both with a Y cord, then the two channels come together at opposite polarity, or 180 degrees out of phase, and they cancel.

Of course the cancellation is not total, and you may even like the result, but it serves to illustrate phase.

[Gainzilla]

Thanks for both the info and your patience. I'm definitely trying to learn as much as I can, but in the meantime I realize a lot of my questions are pretty basic.

The phase makes sense to me now, mostly. Not necessarily why, but at least knowing what the rule of thumb is helps.

I have a couple things going on with my amp that might be a little dicey.
1. How do mosfet stages effect phase? I'm assuming the cathode follower is 0 deg, but what about the fx send stage? Does that flip the phase like a triode stage would?

2. I wired up the low and high input with one jack and a switch. This must mean that the phase going into the PI is different when switched between high and low. Wouldn't that mean one of the two stages will have positive feedback?

I'll try the phase test tonight and see what I can learn.

Thanks again for the pearls of wisdom!

Cheers,

[Tom Phillips]

...How do mosfet stages effect phase? I'm assuming the cathode follower is 0 deg, but what about the fx send stage? Does that flip the phase like a triode stage would?...

It depends on how they are used. The MOSFET itself is just a device like a triode vacuum tube is a device. The way you are using the MOSFET in your FX send circuit there is no phase change. It is a follower circuit with ~5:1 attenuation on the output.

...I wired up the low and high input with one jack and a switch. This must mean that the phase going into the PI is different when switched between high and low. Wouldn't that mean one of the two stages will have positive feedback?

No. It only means that the phase from the input of the amp to the input of the phase inverter changes when you flip the switch. However, the part of the circuit that affects the feedback phase is only the circuitry within the feedback loop. For your amp that is the phase inverter, the power amp and the output transformer. Regarding your amp’s internal negative feedback it does not matter how many times the phase is flipped before the signal enters the phase inverter.

Cheers,
Tom