Plexi distortion sounds like "hash"

[sonofmickel]

I get a super bright "hash" when overdriving the amp. It is not a pleasant sound.
I get the same "hash" when using pedals in front as well(MojoHand DMBL, Tube Screamer etc...) Sounds like sh!t!
Could this be the OT causing hashy brittleness and no girth to the notes?

I have just done another complete rebuild. I used tag-strip for this build using all new components.
The normal channel is super thick and no definition while the bright channel is this. The amp does get loud. I do not get that swirly tube goodness from the amp. very one-dimensional.
It also has a lot of ghost notes.

The circuit is a 1987 with no bright cap on the lead channel volume pot.
Transformers are:
Weber OT W018343(4K Primary, 2-4-8-16 secondary)
Weber PT W022798(720 vct@200ma, 6.3vct@6A), running bias off the rectifier diodes.
Negative feedback resistor is 47k.
Preamp 3 x GT12ax7.
Filtering: 50+50, 33uf, 33uf, 20uf, 16uf.
V1 cathodes resistors are 1k and 3k(did not have 820 or 2k7).
V1 bypass caps are 25uf and .68uf.
V2a cathode R is 1k.

Voltages:
B+ 468, JJ KT77 biased at 35ma.
V1 147, 187
V2 1/7- 168, 6-280, 8-176
P.I. 1-215, 6-212, 2-10, 7-12, 3/8-34

Soundfile:
https://soundcloud.com/emickelson1/plexi-hash-001
https://soundcloud.com/emickelson1/plexi-hash-002

[sonofmickel]

I upped the B+ to 468 using the high voltage taps. The voltages on ALL the tubes are within a few volts of the 1987_MHuss 50W Lead Variant schematic.

The amp sounds a little better still but is not pedal friendly nor is is sounding good with the volume turned up.
There is an ugly distortion(a sand paper/gravel on the bottom of the note) when hitting the wound strings(low E especially) really hard as well. NOT Pete Townsend hard slamming but maybe Bad Company chord blasts. Sound horrible.

[neikeel]

Going to be tricky to help at a distance.

Some pictures would help. Full spec. (transformers, tube types, bias settings etc) ideally.

Then some sound clips if possible (even off your phone) but the top two first.

[Doug H]

there are some minor lead dress issue in the amp, some wires parallel to grids on the pre amp tubes, but judging from some pics I've seen of vintage amps that supposedly sound great, not enough to cause your problems I'd think.

Having said that, straying off the beaten path with even a bit of unproven layout can been trying. This board looks like it was designed for this circuit though so I'd expect that to be vetted (right word?) already.

Shortest possible wires in preamp isn't always the best, those runs to and from the volume pots when they are long and hug the chassis actually act like little capacitors shunting high frequencies to ground and helping to keep the amp stable.

It's hard to see exactly what's going on around the power tubes, where the grids wires are in relation to the OT secondaries and feedback wires can make a difference.

The PI socket wiring, the red grid could be longer to get it away from the cathode wire, the other seems long, but eh...

Still looking, the V1 bright side, again, I wouldn't think it's be the big a problem, but you should get the shorten the plate wire so you can keep it from resting on the heaters. Unfortunately with the orientation of the socket, it kind of has to wrap up there.

it's not like everything has to be perfectly neat and tidy, but if the amp isn't sounding right I'd be suspicious of the location of some of those grid wires.

[sonofmickel]

Thanks Doug!
I will lengthen some of the preamp and P.I. wires and GET THEM CLOSER TO THE CHASSIS!
Speaking of the chassis, it's a weber and is nice and small but maybe too small? Also the tube sockets are all backwards from normal orientation. and it does seem to pose a problem for correct wiring layout, I am not going to redrill for new bolt holes
The board was not designed for this circuit and I mistakenly misplaced the V1 coupling .022 caps(you can see the jumper wires coming off of the plate resistors)that would have cleaned up the P.I. and bias sections of the layout considerably(could have gotten another two-three rows to use, next build!).

The power section is a rats nest!
The OT brown and blue are not twisted, would that help? They are cut to just make the tube sockets. I could solder some extensions and get them twisted if that is a known problem.

I swapped the position of the P.I.100k-82k resistors from the photo to 82k-100k per schematics. Don't know if that makes a difference or not per the .022uf/.1uf P.I. caps.

Question per the Huss 1987 lead variant layout and these caps. Huss has two .022uf caps instead of the typical .002/.1 caps. What difference would that make to amp response?

I'll do all the suggestions! Thanks for the help. I'm getting closer to my toooone(Grand Funk).

[Doug H]

are there any other stand offs for the board other than the preamp end? vibrations could cause problems as well if the board is just suspended by wires.

I wouldn't worry too much about the OT primaries just yet, look at other builds of known good amps with an eye towards the grid wire placement. You don't want much close and parallel to them. The b+ feed usually isn't much of a problem though, so I've been told, read.

I'm not necessarily suggesting lengthening wires, just pointing out that with ad hock home builds one is compelled to follow the matra the shorter the better, and in many ways that's probably great, however that's not how many of the old amps were built and there are subtle nuances to lead dress. YMMV.

btw that red heater wire on V1, the bare metal touches that bolt you can probably say goodbye to your PT.

The bass side of V1 sockets need to be wired cleaner too, the plate and grid connections on the socket look like they are almost touching. Not the issue you are having with lead channel though.

Something else to consider, the bypass mica on the mixer resistors and the tone stack mica, they can be sensitive to heat and reportedly fail quite easily, use a heat sink when soldering or be quick about it, especially with a short lead. I can see what looks like melted wire insulation at some connections which suggests you like to work with a hot iron, a dollar worth of ceramic caps will remove any doubt while you debug the amp.

!st and foremost though, I'd make sure you have nice and neat grid wires to the power tubes that aren't running parallel to the high current on the OT secondaries or the purple feedback wire. It wouldn't hurt to make sure they aren't wrapped around the OT primaries either.

The kind of gravelly overdrive you describe might be a cold joint in the signal wires somewhere too. If you don't like the look of a connection in any of the AC pathways, re do it just to be sure.

[danman]

I'll second what Doug has said. If you look at a lot of older Marshall gut shots, the filament string and blue plate wires to the preamp tubes are laying flat down against the chassis. The yellow (cathode) and green (grid) wires are running up high, above the socket, and away from plate and filament wiring. Whenever a wire has to cross the filament string, try to make sure it crosses at a right angle to it. You would be surprised how big a difference the movement of one wire can make in a build. Check over at http://marstran.com/Marshall%20Pictorial%20Page.htm for an idea of how the wiring should be dressed.

[sonofmickel]

Re-routed some wires, not much difference. Maybe it's a little better?

I thought maybe putting Bright caps would bring the Marshall "kerang" but I was mistaken.
The amp cannot be run with bright caps, ever! They create an unbearable amount of treble.

Uploaded sound cloud files in the OP.

The sound is lifeless and one dimensional played quiet or cranked. To me ,cranked sounds like a half working/bad distortion box. To get my distortion/boost boxes to work, I need to use wildly different settings than on my other amps(Dr Z Maz 18nr, Handbuilt Deluxe reverb, Jet City 22h) while using the same cabinets: 2x12 Avatar with Hellatone 30's, Blankenship 1x12 with Black Shadow 90).

I will rewire the output section next.

Note: I rebuilt the amp before into a JCM800 and the same thing was happening, really bad "hash" tone that matches what this amp does when bright caps are added. I will put a bright cap on the lead channel and record that next so that can be heard.

[Doug H]

There is nothing like bad iron to make an amp sound like ass, but without playing it directly it's not the kind of thing you want to encourage people to suspect. I've blown my share of OTs over the years playing into no load, some went quicker and sounded worse than others, some didn't sound too bad half blown. If cash is no big deal and you don't mind potentially owning a new paper weight then sure, why not rule it out. If you haven't done something to possibly blow the OT then it's probably clutching at straws.

So you play a strat? The lead channel with a bright cap can be pretty hard to tame.

[sonofmickel]

Could it possibly be the PT?

The recordings were done with a 61' reissue SG with 57 Classics.

Cut off little hanging-chads off of resistor/capacitor ends that were sticking out. Rewired some of the Preamp, PI and Power sections. replaced 5.6k for 47k grid resistors - no change in anything. Maybe a little less interference from my articulated ring light(that can't be a bad thing).

Removed ALL 500pf(bright cap, peaking cap). Much better but still having mosquito noise on top of the notes when using dirt boxes. The amps gain sounded much better this way but still could not use a boost box.

Switched the tone stack to 250pf/56k, better tone control (for me). But still had hash/mosquito noise.

Lowered NFB resistor to 33k. Better still.

Replaced .68uf elytic cathode bypass cap with a Solen, much better for mosquito noise, sounds OK with a dirt box!

Still not sure what made the most difference. My Plexi build is now closer to a JTM45 and much more manageable volume wise. It is still missing that open/bloomy tone that I get on friends marshalls(JCM800, 2205).
I will try .047 PI coupling caps next(probably just parallel another .022).
I will try adding back a peaking cap on the bright channel(250p, 500p).
A JTM45 might be what I wanted from the start, a pedal machine.

[Gaz]

I just listened to your clips on my IPhone, and I think you amp sounds 100% normal for a close mic'd Plexi with an SG. You are just hearing phase inverter overdrive. It's "hashy" - it just is. Going from an 800 to a Plexi, I had this same reaction. I was so used to preamp overdrive. Plexis do sound very good in a mix, or mic'd with a smoother sounding ribbon mic. They are special because they breakup like no other big amp does, and have an amazing clean sound all on the same channel, unlike an 800. But if you really hate the OD sound just do one of the millions of mods out there (Mod 5, Brown Eye, Etc), and you will get some smoother more refined preamp OD thrown back in the mix. Jm2c.

EDIT: just read the part about how you modded it t an 800, so maybe I am not hearing the sound you are on my phone, but it sounded normal to me!

[sonofmickel]

Thanks for the reply.

I guess what is really missing is that "oompf" that you should get when hitting a string. That sort of "kerang" is missing as well.
I may have "found" part of the problem...cheap elytic bypass caps. I replaced the cathode bypass with a Solen .68uf and the "hash" sort of disappeared.

I recorded using a Beyer Soundstar MK2, this is a very warm mic, so it probably is rounding off the mosquito and hash. I'll have to remember this mic for that reason. It's a very smooth mic for screechy vocalists.

At any rate the plexi was way to loud to use at any of the low key clubs I play at. I think I will go more towards the JTM45/clean pedal machine.

I would really like to know what Schroeder is doing to get so much open yet warm headroom out of his amps.
New thread coming up!

[Doug H]

I think having a perfectly tight lead channel with a bright guitar is a dream one chases but never realizes. Maybe some of the guys on this board have it going.

IMO the mixer bypass cap Bright cap just seem to aggravate every issue, even if it's healthy. I one problem with the end of a lead from one resistor that was too close to the lead of another component. Maybe the .68 cap was simply too close to the input resistor? I know some of the old boards are packed in there tight and work fine, but it just takes one wrong component to be too close to another one to mess things up.

[Doug H]

pic of your amp was sitting on my desktop so I took another look at. Have you fixed the side of v1 where it looks like the grid wire is laying on top of the plate wire, right at the socket? I also noticed, and not sure if this matters but, one end of your feedback resistor gets pretty darn close to the end the output coupling cap, so basically that's a grid wire. I'd bend the white cap over to the other side of the tag it's attached to.

Anyhow, hope your amp is sounding better. You found a couple issues, a little more clean up might be all it needs?

[neikeel]

I can hear what you mean with the fizziness and hash

What settings do you have in the clips (I could only open second one).

It sounds like a cranked amp that is struggling for power supply with cheap diodes, or sub-par OT..

What happens if you play it with all pots on half with no bright cap?

Your voltages look ok (I would have expected lower voltages with th etones you are getting).

Amps with average OTs will give you those tones and maybe why you are having issues.

I cannot think of anything else other than better lead dress, maybe try some UF diodes (cheaper than new OT) see what you get.

If money is an issue the Classic Tone OT is getting a good rep. (not tried them personally as I try to use originals, Marstran, Metro or ****** in that order for various reasons!)