Modded DR Too Much Filtering?

[thisismyname]

I have a modified DR clone I built 7 years ago when I was 19. I sent it to Gerald Weber soon after to fix some of my noobie mistakes (layout mostly). He recommended upgrading the power filtering in the amp. It currently has two 100uF caps in the totem pole configuration (with the balancing resistors of course) for a total of 50uF compared to the stock 16uFs in parallel for 32uF as I'm sure you all know. He also convinced me to buy an 11 winding interleaved OT that I spent a stupidly ridiculous sum on.

Section B after the choke has a 40uF compared to the stock 16uF.
Sections C and D have the stock 16uF.

My problem is that as long as I've had the amp, it's never sounded particularly good. Kind of sterile and lacking any character. The biggest problem is the onset of distortion. The cleans sound okay, but once the amp starts to distort, it takes on a dull sharpness to the transient that is very unpleasant. It's very hard and jagged, but a dull jaggedness that isn't a high-end thing because removing all the high end doesn't change anything. It's like punch on steroids. I've tried different guitars, speakers and cabs, tubes, and different preamp designs, but to no avail. The only constants in the amp has been the power section, OT, and inherent design of the PI and power amp. I have it currently wired as a Trainwreck Expresso, but with the 3.3M 'verb mixing resistor still in as well as 1k5 in the NFB, and some other gain-reducing ideas. The amp has no reverb or trem circuit, and the normal channel is disconnect before the PI so a mixing resistor isn't needed.

I'm thinking it may have too much filtering because of the lack of compression. 40uF at section B seems really high. Back when the amp was at Kendrick, the tech who actually did the work recommended that cap be lower than what Gerald recommended, but at the time, I blindly trusted anything Gerald said.

Does this sound like a filtering problem?

If so, recommended values? I have four 22uFs, three 16uFs, two 47uFs, and two 100uFs in my stock right now.

I prefer to play in the edge-of-breakup area and using my guitar volume and tone to control everything. I don't need an amp that tries to do high gain.

[xtian]

I'm just guessing, but I don't think it's filtering that makes your edge-of-distortion sounds unpleasant. And the mention that your amp has no tremolo or reverb makes me wonder about the voltages--too high would possibly make it sound bad. Lead dress is another issue. Can you post a voltage survey and gut shots?

[Colossal]

I would consider changing the filtering, yes. But regarding the transition to overdrive, the more abrupt transition that you don't like is part of the nature of fixed bias. I love cathode biased amps for this reason....it's always in between cleanish and moving to dirty depending on your attack, very responsive. You might consider putting a pot in series with the negative feedback resistor, something about 2-5kΩ. They you can dial out a bit of feedback and that will make the amp more lively. Keep going and it will get more aggressive. There are other important things to consider, such as your operating voltages and where the amp is biased.

[Stevem]

Some questions for you.

Where is your bass pot set at ?

What brand output tubes are in the amp?

What are your output tubes biased at , under 16 ma/ .016 amp?

What do you have for a rectifier tube in the amp, a 5U4 or a 5AR4?

Interleaved output transformers are great for clean for Hi Fi type tones ( great frequency response up to beyond 20Khz) and good to max our power transfer, but not for distorted guitar sounds.

A standard replacement OT would help things in my opinion, and so would replacing the PI coupling caps with .022 uf if they are the stock .1 uf in there now!

Also please do this test. Is the harsh tones you dislike taking place during your pick attack, or once the note itself is just gone into ringing?

[thisismyname]

I'll work on a voltage chart, but in the interim, here are some gut shots. I rebuilt the amp about a year ago to allow flying heaters rather than my original poorly done chassis-hugging wires that ran under all the other tube socket wires. I redid all the wires going to the tube sockets, swapped carbon comp for carbon film, and replaced most caps in the process of experimentation. Same distortion problem.

I actually recently added a 100k pot in series with a 1K5 NFB resistor already, but less feedback makes it worse, so maybe a PI, power amp, or transformer thing. I barely even crack the pot if at all.

The bass pot is a 250K, but I guess actually reads 187K.
Mid pot is 100K. I crank it to balance out the boominess. Even when at 1, bass pot doesn't really do anything.

PI coupling cap is 0.02. Originally a .1 as per the Expresso circuit. Felt smooth with a softer response, but too boomy. The connected resistor is a leftover from the trem pot being disabled (Kendrick left the resistor in, but I changed it during gain experiments. Higher values make it too boomy). Going backwards from that junction, the wire goes under the board to connect with the .1 Orange Drop. I'm aware this section might be a little odd.

[thisismyname]

OT tubes are JJs. Tubes were biased at 10 or 11 watts dissapation, don't remember other readings.

The harshness is at the initial pick attack/transient. If I alternate pick while letting the note ring, it's not too bad, but if I stop playing for a few seconds and then start again, the harsh transient is a lot more pronounced on the first note compared to all of the following notes.

[thisismyname]

Rectifier is a JJ 5AR4. I tried a 5Y3 for awhile once. I rebiased of course. The amp felt a little softer, but still was afflicted.

[thisismyname]

How would cathode biasing affect the overall design and functionality of the amp? A simple swap or would I have to tweak a few values elsewhere?

[thisismyname]

I found a voltage chart from a few months ago.

[xtian]

Wow, those voltages are really high. I'd definitely stick with the 5Y3 rectifier.

[thisismyname]

Anything else I could do? Zener diodes? New PT?

[xtian]

Put in the 5Y3 and reevaluate voltages before anything else.

[Phil_S]

Put in the 5Y3 and reevaluate voltages before anything else.

The 5Y3 should give you about 385 on pin 8. I'd guess 6V6 plate voltage about 345. This is the simplest thing to do and well worth trying.

[thisismyname]

It might be a few days before I can evaluate the new voltages.

[pdf64]

Due to various mods, the build looks to have a fair bit more gain than stock; has its unconditional stability, with signal, into an inductive load, been verified?
eg it might be doing this -