Smokin Joe II SRPP and adding voltage meter to VVR

[Fraggy0117]

I'm currently working on a "use only what I already have" amp project, and it's essentially a Weber Smokin Joe II input with a Kalamazoo KEA (https://el34world.com/charts/Schematics/files/Kalamazoo/Kalamazoo_KEA.pdf) power section (only with EL84), and a VVR for the EL84 node.

I chose the Smokin Joe SRPP topology purely because I thought it was weird and I thought it gave a lot of gain - but further research makes me think I was incorrect about the gain. Looking at this article (http://www.valvewizard.co.uk/SRPP_Blencowe.pdf) Merlin states a gain factor of 6 (realizing it's not using the exact values as the Weber schematic, but the ballpark is lower than I thought). This aligns with my build's low volume even turned all the way up.

Before I rewire the input (probably to a cascode), I'm hoping someone can confirm that I shouldn't be expecting a lot of gain from the SRPP - and I've included voltages in my attached schematic in case there's some other issue with it.

My second question is where to add a DC Voltage analog meter within the VVR circuit from this thread: https://el34world.com/Forum/index.php?topic=6899.0. Right now I've got it between lug 2 and the 100K resistor that goes to the gate of the Mosfet. It's not showing any change on the meter, and I'm suspecting I actually need to put it on the "Variable b+ out to cap can" spot. Due to the chassis, that spot is pretty far away, so I'm hoping that's not the only option. The meter is rated up to 1000VDC, and I did test it with a 9v to find the correct positive and negative lug.

The meter isn't on the schematic, but I can add it if it helps.

Thank you!

[sluckey]

Need to know the ratings for the meter, not just that it's rated up to 1000V. Stuff like full scale current would be helpful. Also a hi-rez pic of the meter might reveal a lot.

[Fraggy0117]

Need to know the ratings for the meter, not just that it's rated up to 1000V. Stuff like full scale current would be helpful. Also a hi-rez pic of the meter might reveal a lot.

I looked again and its actually 500v.

I'll see if there's any additional markings not visible in the listing, but this is the meter: https://www.amazon.com/dp/B01HM1QZ4E?psc=1&ref=ppx_yo2ov_dt_b_product_details. It didn't come with any documentation, and I'm having trouble finding a datasheet elsewhere.

[PRR]

No idea why you put that meter inline with the zero-current Gate. You know how it should be.

This stacked triode affair is probably gain of 50 (half Mu).

The Webber schematic runs this in cascade with another dual triode, which not only multiplies gain by about 25, but gives push-pull signal.

Your output stage is NOT push-pull. And you lack Webber's 2nd stage. Yeah, that's gonna be hard work to play. Eat your vitamins, harden your pick.

[bmccowan]

This discussion intrigues me. Not the meter thing, but the Kalamazoo KEA & Weber SJ. I had remembered seeing an Uncle Doug video on the KEA where he described it as a weird form of push-pull. And I found this podcast https://www.fretboardjournal.com/podcasts/the-truth-about-vintage-amps-ep-93/ which does the same. I used to live in Kalamazoo and there is something in the water out there. I'm not sure, but it seems the output transformer is also odd and perhaps needed to get the most out of the circuit. In terms of the SJ, why not use the preamp of the SJ1 which is designed to drive one EL84? If all else fails, you could keep the preamp and wire up the 84s parallel. Or you could keep the same tube sockets and use ECL86s or 6BM8s and go to a more standard push/pull.
Not what you asked - but just throwing some ideas at the wall.

[shooter]

I used to live in Kalamazoo and there is something in the water out there.

I resemble that remark!
let's see, it's in a giant bowl, nations larger producers of paper mill products before sludge dumping was frowned upon, major pharmaceutical manufacturing before toxic waste was frowned upon.  Was the "1/2 way" point between Detroit n Chicago when Alcohol was frowned upon
Ya, there's something in the water, the Kalamazoo River would burst into flames on occasion when I was a kid.

[bmccowan]

I was 8-9 when we lived there 1961-2. Was perfect for that age. We lived within a bicycle ride of the rail yard. My friends and I would race the freights from road crossing to road crossing and then would ride through the yard. We stared straight ahead and kept our mouths shut while the workers waved and shouted hello. We had heard at school that if you talked to them, they would cut your tongue out! And the Detroit Tigers - Rocky Colavito and Norm Cash.

[shooter]

was 3-4 during your time, we lived on the north side Kazoo, clapboard homes, dad couldn't afford wood for the stove, but he could steal all the coal he wanted, my only recall then was getting hauled from the 2nd floor as the chimney was burning, smoke everywhere.  parents said the house was ashes in 15 minutes.  lived around Kzoo till '76, then back in '90, still at the same place, far enough from Kzoo, but 20min and i'm back in the big city 


the plus side, i've worked on many variations of Gibson's from that era

[Fraggy0117]

I'm not sure, but it seems the output transformer is also odd and perhaps needed to get the most out of the circuit. In terms of the SJ, why not use the preamp of the SJ1 which is designed to drive one EL84? If all else fails, you could keep the preamp and wire up the 84s parallel. Or you could keep the same tube sockets and use ECL86s or 6BM8s and go to a more standard push/pull.
Not what you asked - but just throwing some ideas at the wall.

Uncle Doug is where I first learned of the circuit, too. I need to go back and watch that video and listen to the TAVA episode again to see if they mention an uneven Output transformer. I definitely have sound out of mine, just apparently not enough preamp gain to make a good judgement call on if it sounds fine with a "normal" OT.

All good ideas, I'm just a little limited by the chassis which is a small Hammond box - 5"x7" I think. As long as I stick to three 9 pin sockets I'm good to go. Right now the plan is to try an actual cascode design and see how the gain ends up. If that doesn't work then I'll probably do something with ECL86s or cascade the 12ax7 more like a tweed champ and see if that makes a difference.

[Fraggy0117]

Well, parts came in and while I got the meter working the overall output isn't what I think it should be.

First, with the current boosted 12ax7 cascode - supposed to be 230 gain - the output was quiet and also completely distorted even on low volume and 400v going to the power section. I could be assuming wrong, but a cascode directly into a power section would either be: 1. quiet and clean (not enough output to drive power tubes) or 2. loud and distorted (too much gain and overdriving tubes).

So I pulled out the scope and multimeter to check the preamp tube to completely rule out the cascode and in the process maybe found the issue somewhere else.

The B+ at the preamp read 346 (high, I know), with 235v feeding the "top" annode and 159v feeding the "bottom".

For the scope, I'm including the photos since I don't know exactly the correct definitions to use:

For 1KHz input, I got +/- ~1 square at the .1 volt/div setting. After the cascode I got +/- 2 square at the 10 volt/div setting. so that's a .2 volt swing to a 40 volt swing? If I'm reading that right then it's about 200x the voltage, so that seems about right. Now, I am using a benson-style tone switch that drops the output to the volume pot to around +/- 1.2 squares at 10 volt/div but is that enough of a difference to explain the low volume?

Then I went to go check the power tube voltage and perhaps found the real issue. With 402v at the power B+ node - and 400v showing on the meter, there is 4v on the center tap of the OT primary with the tubes in, and about the same on the screen/anode pins for the EL84s. With the power tubes out, It shows approx. the same voltage on the B+ node as the screen/anode pins for the power tubes. I then pulled all of the tubes and put them in another amp and that amp works perfectly - ruling out bad tubes.

What am I missing/not understanding? Is there some internal connection on an EL84 that doesn't exist on a 6v6 that the KEA amp used? I've included pictures of the area if anyone spots a technical issue, but it is crowded and hard to see.

[shooter]

+/- 1.2 squares at 10 volt

make things easier and simply state 12vac, since each square is = 10v, 10X1.2 = 12

400v showing on the meter, there is 4v on the center tap of the OT primary with the tubes in

almost sounds like the sockets are wired wrong
if there's 400VDC without tubes, you add tubes and only get 4vdc somethings sucking the voltage
another option, your ground reference is not the same as the OT CT

[Fraggy0117]

I got the amp running, though without a working meter. But for anyone following along here's the fixes.

The No/Low sound was at least in part caused by the B+ running through the Volt meter. When I short across it I have normal sound and volume. So at this point I think I just have a fundamental misunderstanding of how to add the meter.

I say "in part" because I realized this only after I switched the input to the current boosted cascode circuit (The one rated at 250 gain with a 12ax7). It's possible that the original preamp would've worked but judging how clean the amp is I'm guessing it still would've been low volume.

[shooter]

what are you trying to "indicate" with an analog DC meter?


most all "in-circuit" DCV metering is done via a "shunt" resistor.  the stunt is matched to the meter movement.
the one I use for my off-grid system uses a .05-ohm shunt, the meter then measures the V-drop across the shunt and displays VDC


this allows all the current to pass through without really effecting the voltage AND without getting the resistor hot enough to melt the chassis

[Fraggy0117]

I'm trying to indicate the DC voltage on the output on the VVR where it connects to the OT/Power tubes.

Is the shunt resistor in between the meter and the power node? Does the other side of the meter connect to ground? That makes sense as it's how I'd measure with a Multimeter...

[shooter]

the shunt resistor is in series with the voltage to be measured, the "meter" is matched to the shunt so full defection => the MAX vdc you expect to measure, so;
if your max vdc is 80v, you'd use a meter with 0-100.  THAT IS NOT TYPICALLY VDC, most meter movements are mVOLTS, hence the shunt matching resistor.


I'm talking about an in-circuit meter, NOT a multimeter.  To use a multimeter to measure most vdc you measure from the point you're interested in, like the B+ node at the PS, meter set to VDC, red lead at the B+ node, black lead to chassis ground.


don't probe the PA tubes plate to ground, you can get lots of rabbit holes doing that, not to mention bad things if you're set up wrong, AC is blasting thru the tubes etc.

[PRR]

> I'm trying to indicate the DC voltage on the output on the VVR where it connects to the OT/Power tubes.

Voltage is always between two points. In electronics we usually want "voltage to common", where all loop circuits run together.

Put your Black probe on Chassis and leave it there.

This is same-as a surveyor planting his transit at sea-level and reading the elevations of the top of the hill, the top of the foundation, the level of the road, all relative to one point.

[Fraggy0117]

the shunt resistor is in series with the voltage to be measured, the "meter" is matched to the shunt so full defection => the MAX vdc you expect to measure, so;
if your max vdc is 80v, you'd use a meter with 0-100.  THAT IS NOT TYPICALLY VDC, most meter movements are mVOLTS, hence the shunt matching resistor.


I'm talking about an in-circuit meter, NOT a multimeter.  To use a multimeter to measure most vdc you measure from the point you're interested in, like the B+ node at the PS, meter set to VDC, red lead at the B+ node, black lead to chassis ground.

I'm with you on talking about an in-circuit meter. Thank you for the vocabulary and taking the time to explain this. I'll do some research on how I can figure out a shunt resistor. The meter was cheap and didn't come with documentation so I've been in the dark about it.

don't probe the PA tubes plate to ground, you can get lots of rabbit holes doing that, not to mention bad things if you're set up wrong, AC is blasting thru the tubes etc.

I'm not sure exactly what you mean by rabbit holes, but I've had sparking the first time (on my first amp) I tried to read from the PA plates with my DMM, and haven't tried since.

Do you see an issue with attaching the in-circuit meter to that point, or is it more about an external meter?

[shooter]

DO NOT ATTACH ANYTHING to the PA tube plates except the proper lead from the OT.


get your studies of "monitoring VDC in-circuit" finished, then...


measuring VDC at the PA plates is pretty useless when you have AC signals driving PA tubes, which is sorta the idea of an audio amplifier.
there's some real deep sheet to consider that is way beyond the normal build an amp understanding, stay outta the do-do if you don' want to wind up smelling like farm animals, or charred meters 


"rabbit holes" are distractions created when you don't understand the outcome in the 1st place

[Fraggy0117]

With the audio issue fixed I had buttoned it back up but neglected to share any pictures. Thank you all for your help!