• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

K12G Questions

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

PRR

Member
Joined 2003
Paid Member
Would be good to have links to the items being asked-about.

First look at the PT sold with the S5 K12G and compare with what you have. About twice the iron that S5 thought was acceptable.

The XPWR025, 180VAC, will make 250V DC.

There's little detail about the S5 K12G. The 10GV8 is a 7 Watt tube. Four of them could eat 28 Watts. 240V at 28W is 0.11 Amps. If this was a fancy high power amp we might allow some dynamic current; I'm sure this is a simple amplifier of essentially steady current.

0.2 Amps seems plenty.
 
Thanks PRR.
I searched quite a bit but have found different opinions. Here is a post for example I kinda sorta burned up my K12-G power transformer

that includes the following:

MODEL: K-8G
INPUT: 120VAC 60Hz (BLK-BLK)
OUTPUT: 170VAC 300mA(RED-RED)
5.9-0-5.9VAC 2A(GRN - YEL - GRN)

I just don't know enough to determine if the transformer I have will work as well as the original...my hope was for it to be better. I can't imagine that I would have picked this out without reading about it somewhere but you never know.

Attached is a schematic for reference.

Thanks
John
 

Attachments

  • S5-Electronics-K12G-Tube-Amp-Kit-Schematic~2.png
    S5-Electronics-K12G-Tube-Amp-Kit-Schematic~2.png
    30.5 KB · Views: 253
One thing I notice about your schematic: It shows Pentode Mode output stages. But you said you are using the 1609 transformers in Ultra Linear Mode. It is one or the other (unless you go Triode Wired).

For Ultra Linear and Triode Wired modes, I suggest using four each 100 Ohm resistors (one to each screen connection).

You should have fun listening when you wire one up for the second time.
 
Last edited:
You're right! I just attached it as a reference for the original design. I will be using ultra linear. I think I had the Hammonds hooked up before but only crudely...hard to remember for sure. It sounded better to me in UL vs standard.

What do you think about my current power transformer's capacity?

Where would I add the resistors specifically? I'm green. Also, what is your reasoning for doing so?

Thanks
 
Some pentodes may oscillate in either Ultra Linear mode or in Triode Wired mode.
The 100 Ohm resistors would be connected directly right on the socket screen connections.

UL: The other end of the 100 Ohm resistors would go to the correct Ultra Linear taps (plate and screen of the same 1/2 of the secondary).

Triode Wired: The other end of the 100 Ohm resistors would go to the plate connection of the same tube.
This is an easy one to try if you like experimenting.

The Edcor XPWR025-120 should work OK for your amp.
And I am surprised, it has an 11.6V filament winding.

For everyone who uses a 12.6V transformer filament secondary, but uses four 10GV8 11.6V tubes:
Save your tubes from pre-mature failure
The 10GV8 filament is 11.6V @ 0.45A
Most transformers have 12.6V output, not 11.6V output.
You need 11.6V @ 1.8A (0.45A x 4 = 1.8 A).
12.6V - 11.6V = 1V Drop 0.5V per leg.
0.5V / 1.8A = 0.28 Ohms
(1.8A squared) x 0.28 Ohms = 0.9 Watts heating, use either two 3 Watt or two 5 Watt 0.28 Ohm resistors.
 
Last edited:
Thank you for explaining that. I'll try your suggestions once I get to that point. Sound pretty straightforward.

I know for sure that my transformer is 11.6 on the heater winding at 2A, so I think I'm good there. I think this may be why I picked this transformer originally because it had the spec'd tube heater voltage available. My concern was with the 180V winding at 200 mA capacity. I had read that the original was closer to 300 mA. PRR mentioned that it'd probably be fine but I'm just trying to be sure. I didn't want to move forward with the wiring only to find out that the transformer can't supply the necessary amperage.

John
 
Last edited:

PRR

Member
Joined 2003
Paid Member
Pentode/triode will not be a significant difference to the power transformer.

_I_ think that PT is nearly twice as big as necessary (OTOH the S5 PT may have been marginally small). It is "wrong" in that you maybe could have spent less. However a PT in the hand is always better than any "just-right" PT which has to be custom-wound and shipped heavy-mail.
 
Ok guys, I'm going to take your word for it. I really appreciate your help so far.

I have attached a paint modified schematic.

Can someone explain why the red circled line does not have a coupling capacitor? I would think that the B+ would be feeding into the grid on tube 2A. I would think it would need to be similar to the yellow circle.

I understand that V2A and V4A are being used a phase splitters. Also that V2B, V1B, V4B, and V3B are amplifying. I only think I know this from watching Uncle Doug and Blueglow Electronics on youtube. What I don't know is the function of V1A and V3A.

What I'm getting at is where should I use shielded wire? I'm trying to determine the audio path but I also want to understand where high voltage might exist and prevent the use of sheided wire. I have highlighted in blue what I think it the audio path...for one channel.

Thank you,
John
 

Attachments

  • S5-Electronics-K12G-Tube-Amp-Kit-Schematic~2.png
    S5-Electronics-K12G-Tube-Amp-Kit-Schematic~2.png
    119.8 KB · Views: 199
Last edited:

PRR

Member
Joined 2003
Paid Member
You can figure it out.

What is the likely DC voltage at the plate of the first stage? What is a good DC voltage for the grid of the second stage?

What is your line-in signal level? What signal is needed at the grids to the power tubes?

Where is signal at or below line level, and might need shielding? Where is it amplified-up above line level and likely does not need shielding?
 
JTCamp,

Be careful of shielding.

With a very low impedance signal source driving the input, then when the 100k pot wiper is turned to 50k (worst case condition), the pot wiper output impedance is 25k. This impedance has to drive the capacitance of the shield wire, the grid to cathode capacitance, and the 1+ Miller capacitance of the input tube. This might decrease the high frequency response. This effect could be even worse if the signal source is a high impedance, depending on where the pot wiper is set.

Similar effects are true for the other stages; where plate impedance in parallel with plate load 150k or 47k (and cathode impedance in parallel with 47k) have to drive the shield wire capacitance and grid capacitances.

Short shielded wires that are low capacitance per foot will probably not be a problem; but that requires attention to the distance and shielded wire specifications.

You have a series of 3 areas where the shield wires might have an effect. Suppose the capacitance of each of those 3 areas are -1 dB at 40 kHz. That would be a total of approximately -3 dB at 40kHz, and -1 dB at 20kHz.

I do not use shielded wires on my amplifiers, and I am able to get the output ripple to be less than 500uV, with several amps with less than 100uV. That requires good layout; wiring placement; transformer, choke and OPT orientation and spacing; and good attention to ground loops . . . especially the loop of the first and second cap of the B+; as well as the ground loops of both the input and output connectors. I do not use negative feedback (which sometimes amp designers use when attempting to reduce hum).
 
Last edited:
Hi, This was a popular amp about 8-10 years ago and is still available. It was inexpensive and generally worked OK. There are many mods for it. Most IMO add more to the cost than the performance. Replacement transformers, coupling caps and going to U/L were the best bang for the buck. However even though I don't usually like to send folks to other forums you might look at the projects and posts on the diyaudioprojects.com forum. There are lots of ideas there on these amps.
 
You can figure it out.

What is the likely DC voltage at the plate of the first stage? What is a good DC voltage for the grid of the second stage?

What is your line-in signal level? What signal is needed at the grids to the power tubes?

Where is signal at or below line level, and might need shielding? Where is it amplified-up above line level and likely does not need shielding?
Ok, I'll try to respond to these questions and figure it out. Not sure if I'll get to this week though.

Thank you
John
 
Hi, This was a popular amp about 8-10 years ago and is still available. It was inexpensive and generally worked OK. There are many mods for it. Most IMO add more to the cost than the performance. Replacement transformers, coupling caps and going to U/L were the best bang for the buck. However even though I don't usually like to send folks to other forums you might look at the projects and posts on the diyaudioprojects.com forum. There are lots of ideas there on these amps.
That is around the last time that I worked on it...and I think that's the biggest issue. I didn't know much then but I was familiar with it. Most of the modifications I've already done, but with the chassis change and getting rid of the PCB, I'm having to think about a lot of stuff that I didn't before.

I know I upgraded the coupling caps to Auricaps, I installed 4 UF4007s in place of the existing rectifier, had the Hammonds in UL, and had some snubber caps across the heaters. I think most of it was from "voltseconds" page.

John
 

PRR

Member
Joined 2003
Paid Member
...Not sure if I'll get to this week though....

Take your time. These are good questions. The more you chew the more you understand.

Incorporate 6A3sUMMER's remarks on added capacitance. I once gutted and re-filled an antique oscilloscope. I had the gain-pot on the front panel and the new amplifier on the back, well over a foot away. With pot half-up, treble loss was significant. (The obvious alternative, putting the amp up front, was equally bad, because the final stage of an oscilloscope is high-impedance, especially as I was pushing 99-cent transistors to their 300V 300mA limit to smack a 1936 cathode ray tube.)
 
You can figure it out.

What is the likely DC voltage at the plate of the first stage? What is a good DC voltage for the grid of the second stage?

What is your line-in signal level? What signal is needed at the grids to the power tubes?

Where is signal at or below line level, and might need shielding? Where is it amplified-up above line level and likely does not need shielding?
PRR

Kinda guessing with some of this but here goes. I'm figuring I'll have about 210 volts DC after the RC filtering coming of the diodes. From there I'll drop about 20 volts at R14, leaving me with about 190 volts at the plate of V1A and V3A. Close? I feel like I'm missing something. Confused on calculations for parallel and series resistors.

Assuming the plate of V1A is 190, I think the grid of V2A would also be at 190 volts. Given the different resistors R2 and R4, I would think that during operation this would produce a grid voltage on V2A that was lower than its plate...which is good?

I'm thinking the line in voltage would be around 2 volts AC. I'm not really sure about the amplification power of the 1st triode but I'm pretty sure that the phase inverter provides no gain. Therefore I think that the signal needed at the power tubes would be some product value between 2 VAC and the amplification value that I'm not sure how to calculate.

I'm not sure about your last couple of questions but I'd like to understand. Please let me know what you think. Seems like the only thing below line level would be the input to the first tube, V1A.?

I have learned that the set up I have is called direct coupled. I have also developed a list of other questions as I begin to wire the amp.

Thanks again,
John
 

PRR

Member
Joined 2003
Paid Member
> I'll drop about 20 volts at R14, leaving me with about 190 volts at the plate of V1A and V3A.

Drop in R14 will be less. Even if all the triodes were dead-short, which is not likely. But OK, pencil 190V at C5.

From here to U1a plate we have a 150k resistor. There's probably a voltage drop across this.

As a dart-board guess: the tube and the plate resistor "fight each other", to pull up/down and make audio swing. As a general guide you expect a "fair fight". Tube and resistor split the 190V about equally. With real tubes we may not end up exact equal. 1/3rd to 2/3rd are all typical.

In special cases we work a little off center. And looking ahead, for V2a cathodyne to work "fair fight", with two resistors, we expect cathode at 1/4 and plate at 3/4. So cathode is likely near 47V. And grid will be a tad less, say 46V.

We now have 190V-46V= 144V across 150V. So U1a sucks 1mA, U2a sucks 47V/47K or 1mA. Total suck on R4 is 4mA, 8.8V drop.
 
Ultra Linear, NFB, and Plate Voltage

Guys

I'm trying to make modifications to a S5 K12G amp. I'm going to attach an original schematic as well as one that I modified. I only modified the right channel so you could see the original as well. I'll also include a schematic from Hammond on the PT1609 OPTs that I'm using.

My first question is how do I determine the polarity or phasing of the OPTs? As you can see from my modified schematic, I'm assuming that the polarity is in phase from "top to bottom". What do you'll think?

My second question is about the secondary winding of the OPTs. I understand from Hammond that all of the wires must be engaged to meet specs, but one guy mentioned the bottom winding of the secondary might be a NFB winding...? but that doesn't make sense to me. What do you think?

If I hook up the secondary of the OPT the way I have it drawn, would the NFB connection still go to the + side of the secondary?

Finally, does the change to ultra linear (and the removal of the B+ connection at the screens) change anything downstream on the B+? In otherwords, does the plate voltage of the other tubes change? And does this need to be considered in the conversion to ultra linear?

Thank you very much for any help.

Link to OPTs Hammond Mfg. - "Classic" Push-Pull - Tube Output Transformers - (1608 - 1620, 1645 & 1650 Series)

John
 

Attachments

  • 2.png
    2.png
    30.5 KB · Views: 146
  • S5-Electronics-K12G-Tube-Amp-Kit-Schematic~2.png
    S5-Electronics-K12G-Tube-Amp-Kit-Schematic~2.png
    102.1 KB · Views: 141
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.