New 18W amp: questions

[Lectroid]

Hello all,

I finished wiring my 18W amp today and got it running. Plenty of clean headroom and volume and very little noise.  Nice range of overdriven tones possible with the Volume/MV combination. I'm pleased with it.

I have questions about the tone stack and overall tone.  Schematic and layout below. Hoping someone might have tips for addressing any of these problems.

1. I calculate a plate dissipation of 10.7 watts, ca. 90% of maximum, but the tubes have a glow at their base that I've never seen with 6V6s.  See the picture. Is this red-plating, or just normal for 6BQ5s?

2. Bass pot definitely increases the overall bass but also increases the volume noticeably. The wiring shown on the layout is what I did, but it doesn't look right to me.  Or sound right.    *NOTE: Treble pot seems to 'boost' treble as it should and does not increase volume.

3. The amp's overall tone is too bright, too percussive, too much 'slap' in the notes. Is this what's called 'stiffness?'  My A & B filter caps are a 32-32uF can, with a 15H choke between them.  Would replacing it with a dropping resistor help?  Or just tweak the tone and coupling caps in the signal chain?

4. With the Volume up high for preamp distortion, turning up the Master Volume past a certain point causes the lower notes to fart out.  Blocking distortion, maybe?  How to address?

Much obliged for any and all help, suggestions, opinions!

[Dave]

1. Perfectly normal. You are seeing the filament.


2. The tone controls cut frequencies, they do not boost frequencies. The bass pot cuts a wide swathe of frequencies, so when you turn it up it should get louder.


3. This is for you to experiment with until you like it. Replacing the choke with a resistor might get you where you want to be. You'll just have to try it.


4. Yes, sounds like blocking distortion. It could be addressed by finding a way to reduce preamp gain to where it doesn't swamp out the power amp. Maybe lower gain preamp tubes, or some other method.


Dave

[Lectroid]

@Dave,

Thanks for your comments and for cleaning up some of my confusion.

2. The tone controls cut frequencies, they do not boost frequencies. The bass pot cuts a wide swathe of frequencies, so when you turn it up it should get louder.

Understood.  But the volume increase for this Bass control is much greater than any Fender-type amp I've ever played.  The tone stack design is new to me and I'm hoping someone can spot something I've missed.

4. Yes, sounds like blocking distortion. It could be addressed by finding a way to reduce preamp gain to where it doesn't swamp out the power amp. Maybe lower gain preamp tubes, or some other method.

Would either of these ideas help? 
1. I swapped out the 8.2K grid stoppers for 6.8K because I had them on hand.  Could putting them back to 8.2K help with the blocking distortion? 

2. Or--I could cut back the 2.2M grid leaks to 1M each without hurting the frequency response. Richard Kuehnel mentions this.

Or is blocking distortion just inevitable if I overdrive any power amp past a certain point?

Thanks again!

[AlNewman]

You've seen this tool right?

https://www.guitarscience.net/tsc/marshall.htm#RIN=1300&R1=33k&RT=220k&RB=1M&RM=25k&RL=517k&C1=470p&C2=22n&C3=22n&RB_pot=LogB&RM_pot=Linear&RT_pot=Linear

You could try increasing your 6.8k mid resistor to be more in line with what they have here.  Like a 12k or so.  ?  Might help.

[Lectroid]

You've seen this tool right?

https://www.guitarscience.net/tsc/marshall.htm#RIN=1300&R1=33k&RT=220k&RB=1M&RM=25k&RL=517k&C1=470p&C2=22n&C3=22n&RB_pot=LogB&RM_pot=Linear&RT_pot=Linear

You could try increasing your 6.8k mid resistor to be more in line with what they have here.  Like a 12k or so.  ?  Might help.

Thanks, I have played with TSC before; it's good.  This time it doesn't have a model to fit my simple TB stack.

[WimWalther]

1. Perfectly normal. You are seeing the filament.

Minor correction. These are indirectly heated tubes, so they don't have a filament per se. They have a spiral-wound heater that's inside of the (hollow) cathode.

The visible glowing part is the cathode. If you look carefully, you can  also see the glowing heater leads emerging vertically from the bottom, heading toward their pin connections.

[AlNewman]

"Thanks, I have played with TSC before; it's good.  This time it doesn't have a model to fit my simple TB stack."

In this case, set the mid pot as linear, and in the center it will be half the resistance of the pot value.  So for instance in their default example, 25k set at half would be 12.5K.

[Lectroid]

" TSC doesn't have a model ...."

In this case, set the mid pot as linear, and in the center it will be half the resistance of the pot value.  So for instance in their default example, 25k set at half would be 12.5K.

Great idea, thanks. I' going over there now.

[shooter]

Blocking distortion, maybe?  How to address?

Your best friend here is a scope and sig-gen
throwing parts at guesses is kinda like herding cats, even my border collie gave up 

[Lectroid]

I have the scope. What do I look for?

[shooter]

is blocking distortion just inevitable if I overdrive any power amp past a certain point?

REVEB OFF
AMP warmed up ~ 5min
measure VDC at PA Cathode


I use a 100mVrms input (hot-ish input) set the TS at "5" look at "R" or "RV"
this should give you a clean, amplified signal
check the point between the .022 and "Top" of MV, Both sides


The "ballpark" max clean PRE signal at the PA grid (I call this the drive signal) should be close to the VDC you measured

[Lectroid]

Thank you, shooter,

I should say first that I'm just learning the scope, but I am proficient with the signal generator.   

measure VDC at PA Cathode

Can I assume I measure at the tubes end of the cathode resistor?  This step seems to be one for a DVM rather than the scope(?)  What am I missing?

I use a 100mVrms input (hot-ish input) 

Do you mean the scope input channel?  Setting the voltage scale?  Or..?

check the point between the .022 and "Top" of MV, Both sides. The "ballpark" max clean PRE signal at the PA grid (I call this the drive signal) should be close to the VDC you measure.

Perfect.  I love benchmarks like that.  But supposing the voltages are close to each other?  What does that tell me?  Or what if they're not roughly equal?

Do I need to use the scope to visualize the signal and measure the MAX voltage using the screen grid marks? Or could I just use a MIN-MAX function on the DVM to do the same thing?  (Scope's more fun, sure...    )

Thanks again!

[shooter]

Yes
Sig-gen input to amp, measured at input jack, voltage, I use 880hZ cuz it's less annoying than 1K
Min/Max won't show distortion on a DVM, you want see what you hear

[Lectroid]

shooter, many thanks for all your help on this.  After some morning reading--and re-reading many 18W designs--I got the consensus that NFB shouldn't be used on an amp that generates heavy preamp distortion.  I put one on just as a default since all the Fender amps I've built had one. Never noticed that most of these 18W amps don't have one.  I need a new pair of eyes.      

Thanks again.   

[shooter]

since all the Fender amps

when I 1st showed up here, it was fender this, fender that, fenders everywhere, so I moved to the dark-side stealing from Gibson n Vox


your PA is self-biased, by it's "nature" it really doesn't like being over-driven, it'll love OD pre, OD pedal, just don't expect "Fender" if you OD the PA itself.

[Lectroid]

For this build I wanted something definitely not-Fender.  I'm still tuning the sound but I"m pretty happy with it so far. 

[ac427v]

I wonder if you used the correct option in TSC to model your tone stack. I advocate using the Fender model but substituting your values. The Fender model is the best fit for your circuit because it has a mid resistor--more like the variable resistor of Fender. It does not have the wiring of a potentiometer like a Marshall. Also, your tone stack is plate driven as is the Fender TSC model. Give it a try.

[Lectroid]

TSC works just fine, especially with AlNewman's suggestion to set the Mid pot to a static 6800 ohms.  With the Bass and Treble pots, I can get a fairly flat response across the audio range.

But my problem has always been 'how to tame this beast' rather than 'how to achieve best tone stack possible.'  I am still amazed that two 6BQ5s can put out this much power.  I mean, damn.   

[dude]

The blocking distortion is sometimes common on the 18 watt, some have it, some don't  Try the Paul Ruby mod, it was designed to block this distortion.As far as the overall brightness, 18 watt amps have a tendency to be over bright. Try a brightness "cut" pot, works great on the 18 watt brightness issues. Here's a link, might help, did for me, 

[Lectroid]

@dude,

Thanks for those links.  It's an interesting mod.  My blocking is not terrible, just when I bang the power amp very hard.  As long as I back off a little, it isn't a problem.  I like the 'brightness cut' idea.  I added a bright cap to the initial Gain pot--a mistake.  Definitely not needed on this amp.

[shooter]

sometimes "Brightness" is a roll-off of bass frequencies, I've had pretty good success using 100uF bypass cap on the PA tubes.
with a scope you can sorta prove this.  I feed the amp 80hz, 120hz, 240hz, measure at speaker.  there's a fine-line between letting bass in and muddying up the works



another "tweaker" suggestion, give each PA tube it's own cathode R n C, makes for balancing tubes easier, or unbalancing, there are "no rules" except see smoke, stop! when chasing tone 

[dude]

All l did to my 18 watt TMB was add the Ruby mod, the diode and zener. l used a 12v zener, l think but as Ruby said he used a 6v. My issue sounded just like yours, when the amp was pushed, l’d get this buzzy, blocking distortion..?, other than that sounded great. The Ruby mod did the fix. Also, adding a Vox like bright pot let me dial out the overall brightness the 18 watt usually produces. I highly recommend the bright cut. l saw a schematic and layout on Sluckey’s site, check it out, very simple both Ruby’s zener mod and the bright cut.
I was just postings the link to Ruby’s Zener mod and that video was attached, first time l saw it, pretty interesting, different ways he tried. The schematic in the link for Ruby’s mod is what l did along with the bright cut pot mixed with the treble l got the brightness just right.
Hope this works for you, cheers 😊