Gain stage bypass - query

[NoelW]

There are pro's and con's to muting input or output. Since you only have one additional gain stage I think muting the output might be favorable. One reason is that the power supply for that amp (unless it's modified) operates like phase stages. That can cause problems with any output bleed on the power supply that isn't adequately referenced to 0V (like a filter cap that has a little age on it). This can cause instability. If you would like to mute the input that can work too. As long as the variable attenuation resistor for the drive "channel" is greater than about 220k you could relay ground the grid of the drive stage AHEAD of that resistance rather than relay grounding the output behind the 1M series resistor. Such a series resistance at the input of an additional stage isn't always available (see later info) so grounding at the output is somewhat standard. On that note...

You may want to re think your attenuation strategies. 1M and 2M attenuations for high gain circuits create a SHIT TON of white noise (HISSSSS!!!!) Lowering series resistance in high gain circuits can only help. This typically requires some out of the box thinking and creative alternatives, but it can pay off.

Is grounding the Drive output going to have much effect now that the output will be isolated from the 3rd stage grid, when in Clean mode and vice versa?
Unless there is a risk of capacitive or inductive coupling causing bleed-through of signal/noise to the 3rd stage. ??
I have installed the dropping resistor (33k) and the decoupling cap (33uF) to separate the added V2B gain stage from the others to help avoid bleed-through; is this not enough, you think??

I have had concerns about the HISS also and will change values to suit if there is any. The aux. board makes this easy to do.
I can drop a resistor to ground from the 2M trimpot-Drive pot junction to create a divider and then fiddle the values to suit. This another reason for using the trimpot; I can use it to get a divider ratio and then alter the resistor values to avoid HISS.

My main problem is working within the restrictions of the congested, already populated, poorly made PCB of a 1 channel amp that is pretending to be a 2 channel amp.
I probably should have resisted the owners request to have good overdrive onboard and told him to get a good pedal and use the Clean "Channel".
Too late now, so I will have to make it work.

Other Techs have had success with overdrive mods to these amps, apparently.
Steve Dallman, on the Fender Forum (FDP) has done some, using just the Drive stage "Bypass" method and his clients have been happy with the results.
I was concerned about the Bypass not being sufficient, which lead me to ask you blokes for your advice; and as you and Juan have pointed out; the "Bypass" is not really a proper bypass and will likely cause issues that I don't want.

So on we go.

[Chuck H]

It's not just the V2B drive triode that can be problematic. Notice that the V2A triode is in phase with the input triode (V1A) and fed from the same power supply filter node. This has been a pitfall for this amp model in general. I do think that the branched dropping resistor and cap are suitable for isolating the drive input triode. In fact I think 33k is going to produce lower than desirable voltage, but maybe not.?. Still... Best design would be to operate V1 from one filter node and V2 from another. Since you already created a branched node I would think this is possible. And I would probably just run them in series with a lower resistance than 33k, but you may already have a contrary guru plan that you're trying to follow. Same goes for the channel switching. The method I demonstrated is standard practice. And for good reason. Many have tried to implement channel switching with a simple SPDT relay and failed. SOMETHING always seems to cause bleed in this scenario. Inductive, capacitive or via power supply coupling. If you look at successful commercial designs (which you should have already) you'll see that they ALL ground the drive channel somewhere when it's not in use. You were ready to implement a third relay switch if necessary. Why, now that I've shown you how to do it with two do you want to use only one?

[J M Fahey]

Same goes for the channel switching. The method I demonstrated is standard practice. And for good reason. Many have tried to implement channel switching with a simple SPDT relay and failed. SOMETHING always seems to cause bleed in this scenario. Inductive, capacitive or via power supply coupling. If you look at successful commercial designs (which you should have already) you'll see that they ALL ground the drive channel somewhere when it's not in use. You were ready to implement a third relay switch if necessary.

We are trying to paddle up the waterfall here, we have everything against: VERY high gain , very high impedance, buzzy/aggressive signal (distortion) full of harmonics, constricted space, etc.
All conspire to make buzz jump channels, contaminating the clean signal.

Common complaint is "I can still hear distorted signal in the background".

So you must not only switch between clean and dirty *output* signal, you must also internally mute dirty channel so
distortion isnīt even generated.

Have a look at the cheesy/cheap "splatty" Marshall we are following in another current thread:



Here we have two separate, *parallel* preamps, which are footswitch selectable.

Not just a single triode gain stage in/out of a preamp but two fully separate ones, notice both get signal from the straight from the input jack, and share nothing.

Instead of an SPDT relay selecting between clean and dirty channels they permanently join them through 220k mixing resistors and alternatively ground one preamp output or the other ... in principle the same thing, but, of course, cheaper and simpler.

So "Left" transistor grounds Clean volume wiper to mute it, and "Right" one grounds , not Boost channel Volume but to make it stronger, straight grounds V2a plate and yet it clearly was not enough, so they added a third "contact/switch" to also ground Boost channel Gain pot wiper using the "Center" transistor.

So they are grounding/switching OFF two points on the dirt channel.

So if you have an unused relay section, do something similar.

You might add a transistor switch as a mute similar to the Marshall amp but then you will have to create and add some control line to activate it, not worth it on a single job, so in this case it might be more bpractical to addn a relay gluedn to thye chassis, legs up, and wire it as needed.

BUT, even with output grounded, some dirty splatter

[NoelW]

Hi Chuck & Juan,

I hear you both, loud and clear and I am going to implement your advice, which I always agreed with.
It was just the degree of difficulty and suspect reliability of the traces that lead me to looking for a simpler way to do this.
But life and amp repairs wasn't meant to be easy, but the right way is the best way.

I have spent some time completing most of the other work and I think that adding another relay is the best method to mute the Input and Output of the Drive stage, when in Clean mode.
The closeness and fragility of the traces to the existing relays, concerns me.
I have the components for the added Drive stage mounted on the aux. board, which will enable easy connection to the mute grounding points.

All I have to do is figure out how to wire in the new relay, to the existing circuit, so that the panel and foot switches will operate it as well.
My knowledge of SS electronics isn't great and I am concerned whether the 16V +- power supply can handle another relay. There would then be; two ON and one OFF in one mode and one ON and two OFF in the other.
I have mounted 15W aluminium body resistors externally to a heat sink to replace R85 & R86, which are renowned for burning the PCB.
Blues Deluxe Schematic attached.

I am also going to decouple all the stages from the power supply, to further aid the separation of the gain stages; as per Chuck's advice.
I am thinking that individual dropping resistors and decoupling caps, coming off the "X" Supply node will be simpler to arrange than doing a series drop and cap arrangement, due to the standard layout.
Maybe 10k resistors and 10uF or 33uF caps, both of which I have in small radial sizes, which will make mounting easier.
(Chuck, I only fitted the 33k resistor, so I could easily parallel others to find the best voltage drop, if it was too much. With a measured 330V available, at "X", 33k would have dropped about 30V)

I have also changed the voltage divider 2M resistor at the Drive Input to a 680k with a 220k from the Drive pot top to ground.
It is a bit difficult to change the Volume pot from rheostat to potentiometer, so that will have to stay.

What do you both think of those changes, now.

I really appreciate the advice and help from you both and I don't mind being told that I'm wrong or misguided, when I am.

Cheers, Noel

[NoelW]

Blues deluxe mods update

Hi Guys,

It has been a while since I finished this amp repair and modification and I am posting this update to share what I did and show Chuck and Juan that I followed their advice.
I didn't end up using the Andy Marshman Mods. I started that way, but then went my own way until I got the sound I was looking for.
It was a lot of work, but it turned out very well.
That said, I don't recommend that anyone else do this type of mod to a PCB amp; I will never do it this way again.
It would be far less hassle and quicker, to build a complete new eyelet or turret board and fit a SS Fx Loop and a digital Reverb.

Attached is the Modified Preamp Schematic and some gut shots of the end result.

What I did:
Replaced every electro cap in the amp.
Stiffened the PS with 220uF & 47uF filter caps.
Screwed R85 & R86 to the chassis using 15W aluminium body resistors, to avoid the usual PCB burning from the original setup.
Added three auxiliary boards for: an adjustable Bias, the added Gain Stage using V2B, and the extra decoupling caps & resistors.
Made separate B+ nodes for each preamp stage.
Changed the Volume and Drive Controls from rheostat to potentiometer operation.
Reconfigured the wiring of the Relays to effectively select Clean or Drive and mute the Drive stage in Clean mode, to avoid any bleed-through.
Altered the circuit as per the schematic to achieve a much better Drive sound, while maintaining the desirable Clean sound.
Replaced the Ribbon Cables with individual wires.
Removed C50 and changed C18 to 470pF.
Changed R50, R51 & R52 to 10k, 91k & 110k.
Changed R57 & R58 to 1k.
It is fitted with JJ valves; V1 & V2 - ECC83S, V3 - ECC81, V4 & V5 - 6L6GC.

It works well again, doesn't hum or hiss, has a very good Blues/Rock Distortion sound and it's very nice Clean sound.

Thanks again to the MEF Techs Knowledge Trust,
Noel