Maker Pro
Maker Pro

The Ultimate Tube RIAA-Amplifier?

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Cossor
I am with gorgon on this topic.
To be pedantic, the schematic is probably a circuit diagram and the valves are given V numbers, not T numbers.

My valve book lists the Cossor (capital C) 53KU but does not list a Cossor 5V4, there are other makes for this including Brimar.

The Cossor 53KU is rated at 500V, 250mA
The Brimar 5V4 is rated at 375V, 175mA
Thus the Cossor is far more substantial.

Cossor made other valves, my book lists two or three dozen triodes, double triodes and triode-pentodes.
 

rogerk8

Jul 28, 2011
179
Joined
Jul 28, 2011
Messages
179
Back to business :)

Today I did some elementary measurements on one of the RIAA-amplifiers.

AC-heating (shorted input):

1) 2mVe hum (50Hz only visible) out which equals -48dB re Line (500mVe) which is not that bad
2) Using the potentiometer gave best result while more or less centered. Tuning it to either end made hum increase some 6dB.

Due to 2) a couple of 50 Ohm resistors from both heater ends to ground thus suffices. Even better might be a centre-tapped heater transformer.

At least I know that now :D

So old and primitive solutions may obviously give as good as -50dB hum level relative Line.

Due to no visible 100Hz whatsoever in the output I conclude that old and primitive cossors and passive RC-filtering methodes may give a totally unmeasureable and unaudible 100Hz hum level (the supplied current will however have to be low for this to work without inductors).

More data:

Av(1kHz, 5mV, MM, RIAA)=94~40dB
Av(25Hz, 20kHz, 5mV)=(+16; -20)dB re 1kHz

Note: My primitive old signal generator cannot generate lower frequencies than 25Hz. A rough calculation would yield + 20 log (25/20) i.e +2dB or +18dB total at 20Hz. Which is a little, but not much, low.

I am aware that switching to MC would yield a 5 times worse result but 10mV at PA input is not much when compared to some 1.5V for full output power.

Actually, one should keep in mind that 12,6VAC is supplied to the heaters and only some 2mV is visible at output. This means a remarkable hum rejection of some 76dB!!

This figure is almost comparable to OP's CMRR (>90dB).

Happy New Year!

Best regards, Roger
PS
The first picture has the settings 10mV/DIV and 5ms/DIV.

The second picture has the settings 0,5V/DIV (top) and 2mV/DIV (bottom) at 1kHz.

The slightly higher hum level at the first picture is due to Rs=1k (and not zero).

And please be so kind and note the invisible distortion while using Edison Bias :D
 

Attachments

  • DSCN0561.JPG
    DSCN0561.JPG
    44.6 KB · Views: 148
  • DSCN0559.JPG
    DSCN0559.JPG
    47 KB · Views: 136
Last edited:

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Is it only me that finds an Analog/CRT scope pleasing to the eye? ;)

Chris
 

rogerk8

Jul 28, 2011
179
Joined
Jul 28, 2011
Messages
179
Cossor
I am with gorgon on this topic.
To be pedantic, the schematic is probably a circuit diagram and the valves are given V numbers, not T numbers.

My valve book lists the Cossor (capital C) 53KU but does not list a Cossor 5V4, there are other makes for this including Brimar.

The Cossor 53KU is rated at 500V, 250mA
The Brimar 5V4 is rated at 375V, 175mA
Thus the Cossor is far more substantial.

Cossor made other valves, my book lists two or three dozen triodes, double triodes and triode-pentodes.

Reviewing my attached schematic I now actually agree with you because what it says is actually:

Cossor 53KU, 5V4

There's no colon after cossor and I know Williamson was kind of pedantic about brands. He actually wanted PX25's but preferred Osram KT66's due to ease of heating and similar linearity. I searched my Williamson booklet but can not find it now but I am also certain that he tipped you guys over there to use 807's. Which I have used myself and they sounded great, by the way :)

What I want to say is that I now think you are right. Cossor is just a brand. Please forgive me for my stupidity :)

Best regards, Roger
 

rogerk8

Jul 28, 2011
179
Joined
Jul 28, 2011
Messages
179
Is it only me that finds an Analog/CRT scope pleasing to the eye? ;)

Chris

Wonderful comment, Chris!

But you should see my scope. CH2 is so corroded or something that I had to turn the knob some 20 times back and forth for it to display the signal and I'm of course not even sure it displays right :D

I have to do something about this.

The problem arised after there was some renovation of my apartment due to moisture so my insurance company stored my stuff for some three months somewhere and I can bet it was not in a warm storage...

So I am thinking of using some 5-56 all over the switches and perhaps the potentiometers. What do you think about that?

Maybe WD40 is better because 5-56 is known to desolve rust...:D

Best regards, Roger
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
WD40 is not recognized as an acceptable electronics cleaner. That said I've used it and it's very effective. As WD40 dries over time it becomes tacky though. In fact I've seen it stick metal surfaces together like they were epoxied! So, if you use it I would recommend periodic reapplication or an occasional flush with residue free electronics solvent.

Don't tell my local safety Nazis this but I use mineral spirits as a general solvent for damn near everything, including circuit boards. Hey, I'm also an avid woodworker/ woodturner/ machinist. Mineral spirits is a woodworker's and machinist's best friend. It will remove fresh flux easily. Old flux requires much more work. It's also much less toxic than most other solvents.

Chris
 

rogerk8

Jul 28, 2011
179
Joined
Jul 28, 2011
Messages
179
Today I made some more tests on the RIAA-amps. Here are the results:

Av(MIC, MM, 1kHz, 100mVpp)=70 (7Vpp out)
BW(MIC, MM, 1kHz, 100mVpp)=-2dB@63kHz (where my crappy tone generator ends).

Bandwidth for the MIC-option and therefore the latent RIAA is thus some 80kHz which isn't bad.

Uo(RIAA, MM, 50mVpp)=2Vpp (7Vpp max)
Uo(RIAA, MC, 50mVpp)=10Vpp (max)
Uo(MIC, MM, 150mV)=5Vpp (max)
Uo(MIC, MC, 50mV)=10Vpp (max)

Here "max" means that output is clipping beyond these values. All values are measured at 1kHz.

Using the above test results:

Uo(RIAA, MM, 5mVe)=(5mV*SQRT(2)*2=14,14mVpp)/50*2=0,57Vpp=0,2Ve

Which is a little bit low but equal to the other unit.

Best regards, Roger
PS
Thank you Chris for your tip on WD40. Think I will not use it now. Everybody kind of tells me not to. Leaning towards Teslanol T6. By the way, I think I have localized the problem to CH2 VAR-potentiometer because if it is not fully turned, scope kind of stabalizes.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
PS
Thank you Chris for your tip on WD40. Think I will not use it now. Everybody kind of tells me not to. Leaning towards Teslanol T6. By the way, I think I have localized the problem to CH2 VAR-potentiometer because if it is not fully turned, scope kind of stabalizes.

Just a note here. Living in a subtropical climate I've used many things on my shop machinery to ward off rust. I've used both WD40 and LPS-2 to spray them down. Once, when one of my lathes sat unused for an extended period I was shocked to find that either the WD-40 or the LPS-2 had stuck the tail stock fast to the bed ways. Like it was welded!

The fascinating part about WD-40 or LPS-2 is,.. the way to free things up is to apply more of it! It reconstitutes very well! ;)

Chris
 
Top