18 Watt Valvestorm TMB COMPLETED w/ Voltages and questions!

[ViperDoc]

Sounds absolutely killer to me, nice job!

Many thanks, indeed!

[ViperDoc]

QUESTION!

I'm going to put this amp back on the bench and work on my V1A cathode. I only ever tested about .9 VDC on the pin as I built it, and the target is around 1.2 VDC, yes? I noticed there is a new schematic on Valvestorm (where I got this layout) that is not congruent with the Rev. C (now D) layout.

The V1A cathode bypass resistor/capacitor are specified as 820R and 47 uF on the layout. They are specified on the schematic, however, as 1.5K and 47 uF. The first TMB I ever built specified 820R and 25 uF on the V1A cathode, and that particular input sounds really good as I recall. I'm going to play with both, but being somewhat ignorant about individual triode biasing, what is the effect of the 820R/47 uF bypass cap on the tube/tone vs. the 1.5K/25 uF?

http://valvestorm.com/sites/default/fil ... t_revD.pdf

http://valvestorm.com/sites/default/fil ... B_revD.pdf

Thanks a bunch.

[Bieworm]

QUESTION!

I'm going to put this amp back on the bench and work on my V1A cathode. I only ever tested about .9 VDC on the pin as I built it, and the target is around 1.2 VDC, yes? I noticed there is a new schematic on Valvestorm (where I got this layout) that is not congruent with the Rev. C (now D) layout.

The V1A cathode bypass resistor/capacitor are specified as 820R and 47 uF on the layout. They are specified on the schematic, however, as 1.5K and 47 uF. The first TMB I ever built specified 820R and 25 uF on the V1A cathode, and that particular input sounds really good as I recall. I'm going to play with both, but being somewhat ignorant about individual triode biasing, what is the effect of the 820R/47 uF bypass cap on the tube/tone vs. the 1.5K/25 uF?

http://valvestorm.com/sites/default/fil ... t_revD.pdf

http://valvestorm.com/sites/default/fil ... B_revD.pdf

Thanks a bunch.

Try it. Changing the bias can alter the sound
... if it were me, I'd just try it out

[colossal]

The V1A cathode bypass resistor/capacitor are specified as 820R and 47 uF on the layout. They are specified on the schematic, however, as 1.5K and 47 uF. The first TMB I ever built specified 820R and 25 uF on the V1A cathode, and that particular input sounds really good as I recall. I'm going to play with both, but being somewhat ignorant about individual triode biasing, what is the effect of the 820R/47 uF bypass cap on the tube/tone vs. the 1.5K/25 uF?

http://valvestorm.com/sites/default/fil ... t_revD.pdf

http://valvestorm.com/sites/default/fil ... B_revD.pdf

Thanks a bunch.

Note that for the Valvestorm Normal channel, a single triode is used. To maintain the voicing of the traditional 18W Normal Channel, the cathode resistor value should be doubled, and the capacitor value halved. Therefore for a single triode, that would be 1k5 and 25uF. Valvestorm's schematic is correct, but their layout is wrong. For an actual (model 1974) 18W, Lite IIb, or Modern Classic, a single tube is used with both triodes running a shared bias resistor and capacitor, and plate load resistor. The resistor is halved to 820R and the capacitor doubled to 50uF (or 47uF). The plate load resistor of the original, paralleled triode is 100k, and this is equivalent to 200k if used for a single triode with the same voicing, belying Marshall's propensity to co-opt amp designs, from Vox in this case. You can always try both voicings and see what you like. If you want max gain and a warm bias, go with 100k/820R/25uF. I would be inclined to reduce the 25uF to 2.2-4.7uF to reduce and keep the bass from sounding like a tuba.

[ViperDoc]

Awesome pearl there, my brother. I will be testing that out for sure.

Just a bit ago I pulled a 25 uF cap and swapped it in for the 50 and kept the 820R in place. It solved some dirty issues I was having, but was a bit weak-sounding. So I pulled one leg of the 820 and slapped on a 1.2K resistor and it sounded awesome! Super clean, dynamic. And then I, of course, made it worse. When I snipped out the 820 and tried to apply some Colossal chop sauce onto my new resistor wiring (I didn't come close, don't worry), somehow I created a ticking sound when I fired it back up. So I cooled it back down, wicked off the existing solder, tightened the leads around the turrets and re-soldered the 1.2k and now the ticking is gone, but now the overall tone sounds darker somewhat for reasons I cannot explain. It sounds good, but not quite like before.

I'm going to try the 1.5k with the 25uF (or some other lower value if I can find one) and return and report. Thanks for the help.

[JMPGuitars]

Awesome pearl there, my brother. I will be testing that out for sure.

Just a bit ago I pulled a 25 uF cap and swapped it in for the 50 and kept the 820R in place. It solved some dirty issues I was having, but was a bit weak-sounding. So I pulled one leg of the 820 and slapped on a 1.2K resistor and it sounded awesome! Super clean, dynamic. And then I, of course, made it worse. When I snipped out the 820 and tried to apply some Colossal chop sauce onto my new resistor wiring (I didn't come close, don't worry), somehow I created a ticking sound when I fired it back up. So I cooled it back down, wicked off the existing solder, tightened the leads around the turrets and re-soldered the 1.2k and now the ticking is gone, but now the overall tone sounds darker somewhat for reasons I cannot explain. It sounds good, but not quite like before.

I'm going to try the 1.5k with the 25uF (or some other lower value if I can find one) and return and report. Thanks for the help.

Both the resistor and capacitor affect the bass tone in the amp. High values increase the bass, which is one of the reasons the cap gets lowered with a higher resistor value. You could try 1.2 to 2.2K for the resistor, but as Colossal said, you may want to reduce the capacitor significantly.

[ViperDoc]

That’s interesting to know. So far I’ve been diming the bass to get enough. We’ll see how much is on tap with a 1.5k.

[JMPGuitars]

That’s interesting to know. So far I’ve been diming the bass to get enough. We’ll see how much is on tap with a 1.5k.

I don't know if it will be that significant. You can change a cap or two in your tone stack to increase the bass.

[ViperDoc]

WELL!

I took the V1A 1.2K out (my initial substitute for the 820R) and put in a 1.5K with the 25 uF cathode bypass cap (Colossal's suggestion for the 50 uF substitute). It officially sounds amazing. AMAZING. I had some ringing in my amp prior to these recent mods and now it appears to be gone. Perhaps I did some bad soldering, I don't know. When I removed the 50 uF V1 bypass cap it disappeared.

The voltage on the V1A cathode pin is now 1.28 VDC. It sounds killer.

I mentioned previously I had the bass dimed, but that is on the TMB channel only, of course. The normal channel has a wide range of tones on the tone knob, and I prefer it on the higher end. It is a bit flabby and dull when rolled off.

I'm going to study the other voltages and see if I want to do more.

All of your help is much appreciated!

[ViperDoc]

I thought I'd post my latest voltages. I apologize for not using the sheet you've created, but when I try and import it into an application that can enter numbers, it becomes very difficult to read and life is short, man! If I knew how to create an "editable" version of your voltage sheet, I would. Maybe someone knows how to do that.

I noticed today with extensive playing, the tone becomes flabbier with time. Is this normal?

Here you go:

V1: normal 12AX7
pin 1 = 166 VDC
pin 3 = 1.29 VDC
pin 6 = 185 VDC
pin 8 = 1.82 VDC

V2: tmb 12AX7
pin 1 = 162 VDC
pin 3 = 1.21 VDC
pin 6 = 257 VDC (I don't have a good reference for the TMB 2nd triode plate)
pin 7 = 163 VDC
pin 8 = 163 VDC

V3: phase inverter 12AX7
pin 1 = 198 VDC
pin 2 = 51 VDC
pin 3 = 73 VDC
pin 6 = 202 VDC
pin 7 = 51 VDC
pin 8 = 73 VDC

V4: EL84
pin 2 = 32 mVDC
pin 3 = 11.2 VDC
pin 7 = 341
pin 9 = 320

V5: EL84
pin 2 = 31 mVDC
pin 3 = 11.2 VDC
pin 7 = 341
pin 9 = 320

V6: EZ81
pin 1 = 298
pin 3 = 353
pin 7 = 298

Power Rail:
point A: 355
point B: 324
point C: 274
point D: 261

[Daviedawg]

"the tone becomes flabbier with time. Is this normal?"

Not in my experience. And my thought is that the tone should not change after the amp has warmed up. Everything should be stable.

Is it your listening experience after a long period of playing? Or have you accounted for that?

Dd

[ViperDoc]

Ear fatigue could be a factor.

I've been looking at various adjustments like upping my cathode bias resistor value to get closer to 12 V on the PT cathodes. Also, I may adjust my PI cathode bias resistor value, though I'm not sure which direction will clean it up. My research suggests that if I decrease the PI cathode bias resistor value, I lower PI/preamp distortion in favor of PT distortion. The third is to lower the PT grid leak resistors. I was reading Blencoe's explanation of it all and there seems to be quite a bit to it.

Basically what I'd prefer is a strong, slightly cleaner tone. I have it at lower volumes, but when I turn this amp up it gets angry! It sounds really good, I suppose I'm interested in a louder JTM45-style, blues breaker dirty/clean tone. I also read up on 7189 tubes, seemingly a more durable choice for higher plate voltages. Maybe that's worth looking at, though I'm uncertain how I would need to change my components. The pinout looks to be the same, unless it's the A version. No smoke allowed!

[JMPGuitars]

Ear fatigue could be a factor.

I've been looking at various adjustments like upping my cathode bias resistor value to get closer to 12 V on the PT cathodes. Also, I may adjust my PI cathode bias resistor value, though I'm not sure which direction will clean it up. My research suggests that if I decrease the PI cathode bias resistor value, I lower PI/preamp distortion in favor of PT distortion. The third is to lower the PT grid leak resistors. I was reading Blencoe's explanation of it all and there seems to be quite a bit to it.

Basically what I'd prefer is a strong, slightly cleaner tone. I have it at lower volumes, but when I turn this amp up it gets angry! It sounds really good, I suppose I'm interested in a louder JTM45-style, blues breaker dirty/clean tone. I also read up on 7189 tubes, seemingly a more durable choice for higher plate voltages. Maybe that's worth looking at, though I'm uncertain how I would need to change my components. The pinout looks to be the same, unless it's the A version. No smoke allowed!

7189s don't need any changes, they work the same as EL84s, but can handle higher voltage (in theory, but EL84s also handle higher voltage with proper screen voltages). You can read more on the subject here: viewtopic.php?t=4291

Now you can increase the output/headroom slightly using the 7189s, but changes aren't necessary.

There's no harm in trying different grid reference resistors for the power tubes. Try 330K or 220K and see how you like the sound. I've run all those, the 470Ks, and what I call the "stupid simple master volume" to sweep through a bunch of different values.

Your phase inverter plate voltage is a little low. I usually like to see around 225V there.

Before you change anything with the PI, I would check your output dissipation. Maybe try biasing for 10 or 11W per tube (but stay under 12), and try the 220K or 330K grid reference resistors. That combination might give you closer to what you're looking for. Just be aware that running the tubes hotter may introduce noise (but less likely an issue with 7189s).

BTW- if the tone is flabby after a period, and you find out it isn't ear fatigue, it could be a bad tube or an issue with something else. I'd try a different set of known good tubes, and see if you get the same results.

Thanks,
Josh

PS. If you do get 7189s, post a demo. I've been interested in getting some, but don't want to spend the money without hearing them first.

[ViperDoc]

"Before you change anything with the PI, I would check your output dissipation. Maybe try biasing for 10 or 11W per tube (but stay under 12)"

Thanks, Josh. What's the best way to do that? (EDIT: I just found a cool calculator for that. Time to gather more data after work. )

[ViperDoc]

DELETED

[crgfrench]

The 470k is on V2. Schem & layout look OK and agree on V1 to me.

[ViperDoc]

You are correct, I boneheaded that last “finding”. Next!

[ViperDoc]

OUTPUT DISSIPATION CHECK:

I used the methods described on RobRobinette.com to calculate some numbers for my power tubes using measurements from my powered-up chassis. Some very interesting data here.

First I used the basic calculator:
PT plate-to-cathode voltage = 329 VDC
Cathode resistor value = 125R (verified with DMM)
Calculated per EL84 dissipation = ulp...14 WATTS.

Secondly, I measured the voltage drop from the OT CT to each PT plate and found each at 13 VDC. The Resistance between the OT CT and the V5 plate was 312 ohms and to the V4 plate was 329 ohms. Using Ohm's law I calculated my V4 plate current as 13 VDC/312R =39.5 mA and 13 VDC/329R = 40.6 mA for V5. Again using Ohm's law with an average of 40 mA of plate current, I have an average PT plate dissipation of 13.2 Watts or 109.7% dissipation.

SO! Assuming this is all correct, what would you recommend? Previous recommendations included:

1) lowering PT grid reference resistor values
2) EDIT:[lowering] "CORRECTED TO" RAISING PT cathode resistor values
3) installing 7189 power tubes (I don't want to do this until I correct my voltages and power for proper EL84 operation).

Where do you think you would start first?

Thanks.

[JMPGuitars]

1) lowering PT grid reference resistor values
2) lowering PT cathode resistor values
3) installing 7189 power tubes (I don't want to do this until I correct my voltages and power for proper EL84 operation).

Where do you think you would start first?

Thanks.

No, RAISE the power tube cathode resistor value. Lowering the value will increase the output. You should be experimenting between 150 to 200 ohms, with 180 ohms being the best place to start.

That's what you do first. The grid reference resistors are more about changing your distortion/headroom sound. Biasing is done with the cathode resistor...that's why it's called cathode bias.

...and even if you did install 7189 tubes, you'd be about 100% dissipation there anyway.

Thanks,
Josh

[geoff 1965]

like Josh said 180R would be a good starting point,this will bring the dissipation down and also bring your cathode voltage up closer to 12V.
when you measure your voltage drop "centre tap to plate OT" leave the multimeter on for a good 5 mins to allow the tube to heat up and stabilize,with the unequal resistance you should have a slight difference in voltage drop for each tube.use the 20 or 200 DCV range to get an accurate reading "noticed you have 13V drop on each tube"