After GLOW ISSUE

[deepgroovebiz]

Afterglow issue
The music breaks up at louder volumes on 95db and 92db speakers
I am not sure if I am missing something

I put all new filter caps in, because it has the deyoung lower voltage transformer, I had no need for higher voltage caps
I put a turn on delay relay in
The amp uses 12at7 and a 2a3 featuring the tfa 204 transformers

2a3 3k cathode resistor
B+ before filter 430V
B+ at plate of 2a3 394V
Bias voltage of cathode of 2a3 253V
Bias volatage of 12at7 (using one section) 1.6v
Plate voltage of 12at7 235V
Voltage at ccs 251V

Thanks
Chas

 

[Paul Birkeland]

Yes, your amps are not operating properly.

How old are the 2A3s? 

How do the 3K cathode resistors measure? 

The output of the C4S (plate of the driver tube), as far as I can see from the manual at least, is supposed to be 100V.  You have 235V instead, which isn't so stellar.

I note that the original manual calls for a 5965 triode driver.  You have 1.6V of bias over a 274 Ohm cathode resistors, which means you are running about 6mA of plate current.  6mA/1.6V of bias on the 12AT7 curves puts you up at 175V, 6mA/1.6V of bias on the 5965 curves puts you down at around 120V. (The manual calls for 1.4V of bias and 5mA, which nudges you down a little lower.

My suggestion is to buy a pair of 5965 tubes and recheck your voltages, or redo your amps to use the 12AT7 instead.

-PB

(PS - I also have new chassis plates for Paraglows from a small mistake while attempting to only order one pair)

 

[Doc B]

Yeah 12AT7s would be part of the problem. That was a long time ago, but IIRC the 5965 plate voltage might end up running a little higher than 100V, as PB suggests.

As for the music breaking up at higher volumes, the directly coupled Afterglow puts out around 3 watts. It's forte is delicacy over grunt. Not going to play a 92dB speaker super loud without clipping, even with the voltages properly adjusted. It's going to be more happy with speakers of 97dB sensitivity and higher. This is one distinction between that older design and the Stereomour II, which with 4 cap coupled watts sounds surprisingly ample on a 94dB Jaeger.

 

[deepgroovebiz]

I bought them from MIke L years ago, they came with 12at7 the 2a3 are new and tested
So I put the 5965 tubes in and I am getting 1.6v 164 on the plate

thanks

 

[deepgroovebiz]

I also have iron for paraglows, I had new plates made per original pattern but I like what you did better
How much do you want for the plates
thanks
Chas

 

[Paul Birkeland]

You could try trimming the 100 Ohm resistor on the C4S board to reduce the driver current a little bit and pull the plate voltage down. 

I would imagine that PJ's and Doc B's memories will have some recollections of the variations on the amp and what voltages may or may not be expected. 

-PB

 

[deepgroovebiz]

Thanks it is really quite well
No redness on plates

How much do you want for the extra plates?
Are they 12x12?

 

[Paul Birkeland]

You can PM me about the plates, for some reason I can't PM you.

There isn't an issue here of red plates vs. no red plates.

If the plate voltage of the driver tube is too high, then you will run out of voltage in the output stage.  This is what caused your original issue.  This condition will also cause the 2A3 to draw excessive current, which will eventually damage components in the amplifiers.

If the plate voltage of the driver tube is too low, you'll run out of current (instead of voltage) on the 2A3. 

This is why the driver plate voltage on a directly coupled amp is so important. 

 

[deepgroovebiz]

thanks
I have no dc coupled amp experience

deepgroove1.com you can get ahold of me here

 

[Paul Joppa]

Been a long time!

I have few notes, but I do have two hard-copy manuals; both have the same copyright tear (1997); penciled notes on the front pages indicate one of them is "second version with corrections."

In the original design (by John Tucker) the C4S draws its power from the 2A3 cathode (center pin of the 50-ohm hum pot) and is set for 8.6mA. The driver is a 5965 with the two sections paralleled. The 182-ohm unbypassed cathode resistor drops 1.56 volts; this combination gives about 100v at the driver plates, which are connected to the 2A3 grid (pin A3, connected to terminal 9). For some reason the voltage check at terminal 9 calls for 264 volts, but it should be 145 volts (the 2A3 filament pins A1 and A4 call out the correct 145 volts).

There was no grid stopper between the driver plate and the 2A3 grid, but I highly recommend adding one, right at the A3 socket lug.

As shown in the circuit diagram, the 2A3 operates at 250v plate to cathode. The cathode voltage should be 145 volts as specified in the voltage checks. In the earliest version, the cathode resistor was 2500 ohms, so the total current is 58mA through the resistor plus 8.6mA through the driver plates, plus whatever the C4S bias current is - probably around 2mA. Total 68.6, well over the tube's rating. Presumably for that reason, the revision switched to a 3000 ohm resistor to get about 58mA total, 48mA through the 3K resistor. That resistor then dissipates 7 watts and is rated 10 watts - there is not enough safety/reliability margin; it should be at least a 25-watt part. Note that, with 253 volts as you have measured across the 3K resistor, the current is 84mA and the resistor dissipates 21 watts!

After the original Afterglow, there was a period of rapid change and confusion, as the ParaGlow replaced the Afterglow, and the deYoung transformer became difficult to get so we transitioned to the Magnequest PGP8.1 which had significantly higher voltages. We also switched to using a single section of the 5965, and eventually the Russian 6N1P for the ParaGlow. My records and memory are insufficient for tracking that rather involved history.

I hope that's at least a little help.

 

[deepgroovebiz]

Using the 5965 brought the voltage down to 195 at the cathode of the 2a3
I am going to try paralleling the sections of the 5965 which will draw more current pulling down the voltage even more

I have been busy transitioning, sorry daddies I am into women, the entire month of August was spent in thailand for surgery and I was dragging all of september


https://www.pof.com/viewprofile.aspx?profile_id=132512830

thanks
Chastity

 

[Doc B]

My memory of that circuit is pretty foggy. I do remember JT and I penciled it out on a napkin at a Chinese restaurant in Vegas at CES. If the sections of the 5965 are not paralleled I think you probably have something that was modified to a later version. You probably need to change the small (current set) resistor on the C4S board as well as the other changes PJ called out for parallel operation.
Best of luck on your transition. Recovering from surgery is always a longer process than we want it to be.

 

[deepgroovebiz]

Thanks sweetie
I bought them and a bunch of other amps  from MIkey back when he was Bristol street
So who knows what was done to them

 

[pboser]

Doc, would it be inappropriate to ask if PJ would post the latter version schematic somewhere?  I have a pair of Afterglows, converted from original SEX amps!  (Original owner -do you have records that old?  ) They could use some attention so it makes sense to update them too.

Thanks for everything,
Pete

(Hoping to get to Capital Audio Fest this weekend, and to run into JRob, Roscoe, Blackie and some other NYNoize attendees!)

 

[Doc B]

I don't know that a later version schematic will be that close to what we would do to one that came into the lab for a refurb these days.  When we service Afterglows and Paraglows the most important thing is to replace the power supply caps. They were running right at the limit of the 450V volt rating and are spent after all these years. We replace them with two 350V rated caps in series and will replace the rectifier diodes just for extra insurance. And as PJ says the 2A3 cathode resistor should be replaced with a 3K ohm 25W resistor. Other than that I don't think the changes we may have done really had that much impact. The 6N1P is pretty nice, but they don't last. And these days I think the 5965 is more linear and sounds a little better.

Cripes, it's snowing here. Winter is beset upon us.

 

[pboser]

OK, thanks.  Those are good suggestions I will implement.

Keep the snow for a while - don't send it east just yet.

 

[Paul Joppa]

Both of the manuals I have use the same circuit diagram, which shows an obsolete current source design and 2500 ohm cathode resistor. In the manual marked "Second version with corrections" the parts list and instructions show a 3K resistor, but the circuit was not updated. It also shows a 100K resistor across the input jack, but the text and the parts list show 249K. There may be other inaccuracies - these are just the errors I noticed yesterday and today. As far as I know, there was never a more up-to-date accurate schematic.

 

[pboser]

Thank you Paul.  I will change the cathode resistor and the PS caps, replace the diodes and be satisfied.  Oh, how about the grid stopper- what value should that be?

Thanks,
Pete

P.S. As I drive through Maryland toward CAF tomorrow I will pass signs for Joppa, MD.  https://en.wikipedia.org/wiki/Joppa,_Maryland

 

[Paul Joppa]

Grid stopper can be anywhere from 100 to 10,000 ohms; we've used 220 ohms pretty much everywhere. Carbon composition is widely regarded as the best, but metal or carbon film are acceptable at lower resistances (under 1000 ohms?)

 

[pboser]

Thank you Paul! 

And thanks to Chastity for bringing up the old amps.