Author Topic: V-M 557 Plate/Screen Grid Voltage Questions  (Read 1469 times)

HiFiFun

  • Full Member
  • ***
  • Posts: 149
    • View Profile
V-M 557 Plate/Screen Grid Voltage Questions
« on: January 12, 2019, 08:10:10 PM »
Hello all,
I was about to re-assemble this nice tabletop V-M with the
ultralinear amp. It has been recapped and bad resistors
replaced.

The 6BQ5 power tubes "should have" 265 and 270 volts on
the plate and screen grid respectively, per the schematic.

I measure 330 and 337 volts on my amp.
My household VAC is 122 volts, and I did the voltage measure
with a DMM.

The Sams data was calculated using a VTVM. I have done some
research and found that a modern DMM will give higher readings
due to impedance design differences in these 2 meters.

I am in no way fluent in electronic circuit design and
troubleshooting, so can someone give an opinion if these
voltages are wnl?
I know different versions of 6BQ5,  or e.g. 7189, can have
higher voltage specifications.

The schematic is in the downloads section.
Thank you,
HFF

electra225

  • Hero Member
  • *****
  • Posts: 3221
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #1 on: January 12, 2019, 09:04:18 PM »
Sams may specify 117 volts line voltage and a VTVM with 20,000 ohms per volt sensitivity.  I would say at 122 volts line voltage and by using a DMM, the readings you got are about normal.  If I wanted to be sure, a suggestion would be to operate the device on a Variac, set to the voltage specified in Sams, and use a VTVM to make the measurement.  I don't like using digital meters for reading voltage on the old tube equipment.  I like digital for measuring resistors, but that is about all I use them for on the old tube stuff.  Good luck.
I don't need Google.  My wife says she knows everything.

TC Chris

  • Hero Member
  • *****
  • Posts: 2034
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #2 on: January 12, 2019, 10:00:36 PM »
Sams may specify 117 volts line voltage and a VTVM with 20,000 ohms per volt sensitivity. 

Wait... do you mean VOM?  I thought a VTVM has higher impedance.  Not sure....

Chris Campbell

TC Chris

  • Hero Member
  • *****
  • Posts: 2034
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #3 on: January 12, 2019, 10:06:29 PM »
My 566 schematic, the later one with the tapped output transformer primary, shows 280 V at plate and screen grid, and 285V at the OPT center tap.  It doesn't specify input voltage but the power transformer secondary voltage is 310VAC.

Chris Campbell

electra225

  • Hero Member
  • *****
  • Posts: 3221
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #4 on: January 13, 2019, 07:31:45 PM »
A VTVM does have high impedence.  I never mentioned a VOM.  What am I missing?
I don't need Google.  My wife says she knows everything.

voxACthirtee

  • Full Member
  • ***
  • Posts: 159
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #5 on: January 13, 2019, 08:01:58 PM »
A couple things.1) Why are you referring to this amp as "ultralinear"? Ultralinear would have a grid voltage far lower than you listed. Does it just have grid taps for the EL84? That alone does not make it ultralinear.........
2) The final/actual plate voltage of any power tube is essentially "set" by the bias.    Check the value of the bias resistor or resistors on the EL84. It may be out of
    spec.........If you have ever had an adjustable bias amp and had to set the bias, the plate voltage will also vary a bit as you change the bias control higher or lower......... 3)AND, since a transformer is simply iron wound at different ratios, if you are at 122v, you will get a higher overall output on the secondaries, but alone won't give you as much as you are showing. Would be part of the reason though.

electra225

  • Hero Member
  • *****
  • Posts: 3221
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #6 on: January 13, 2019, 09:02:16 PM »
The plate current is set by the bias.  The plate voltage may vary a few volts as plate current increases or decreases.  The purpose of setting the bias is to find the optimum setting for controlling current in the output tubes or transformer.  Too much bias "hot bias" could damage the output tubes or transformer.  I was not aware of bias having any consideration whether an amp was ultralinear.  I always thought an amp with tapped output transformer for screen voltage was an ultralinear amp. 
I don't need Google.  My wife says she knows everything.

TC Chris

  • Hero Member
  • *****
  • Posts: 2034
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #7 on: January 13, 2019, 09:35:23 PM »
Me too.

Here's what Wikipedia has to say about it, which seems to apply to the V-M circuit.
https://en.wikipedia.org/wiki/Ultra-linear

Here's a 1953 article describing it.
http://www.r-type.org/articles/art-115.htm

And here's an extended discussion.  Toward the bottom there's a "Table 1" showing EL84 and EL34  characteristics.  It shows equal plate and screen grid voltages for the EL84/6BQ5.
http://www.oestex.com/tubes/ul.html

Chris Campbell


electra225

  • Hero Member
  • *****
  • Posts: 3221
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #8 on: January 13, 2019, 11:25:28 PM »
There appears to be two different schematics on this forum for a model 557.  One is in Sams 452, folder 16 and one is in Sams 414, folder 10.  One appears to have an ultralinear amp, and one does not.  Am I having a senior moment here?  Sams 452 has the 557-A schematic, Sams 414 has simply model 557.  The 557 schematic shows the ultralinear amp, while the 557-A schematic does not show the ultralinear amp.  The schematic for the 557 only shows 240 volts on the plate of the 6BQ5.   :-[
I don't need Google.  My wife says she knows everything.

voxACthirtee

  • Full Member
  • ***
  • Posts: 159
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #9 on: January 14, 2019, 11:25:08 AM »
Its semantics but i used the "set" for voltage in quotes because the voltage will move as you turn the bias pot. And the final plate voltage will vary, based on your bias setting. Nothing i said was incorrect. And often times its not "just a little".Would really depend on if the tubes themselves are "hot or cold" and what dissipation you are setting them for. Now, it really shouldn't vary 40+ volts on what i assume is a cathode bias. But a bias resistor thats way off can obviously effect final plate voltage. Also, an aside, a lot of EL84 designs i've come across will leave the grid voltage really close to plate. However, under load they settle in lower as they should. So a static no load voltage test doesn't always tell you want you want it to. Also, no, screen taps are screen taps. Its said by folks i've talked to, that they are "better" as far as designs go, but they fall into the "gets expensive to implement" category. A couple of 470ohm or 1K resistors are apparently a whole lot cheaper than an additional couple of transformer taps. And on top of that "ultralinear" is a term that is used to describe a particular design where the screen voltages are around 43% lower than the plate(higher for some tubes) I could google it, but so could anyone, and the guy ran the tests, formulated the percentages for each tube where the lowest distortion figure was by lowering the grid voltage that much, while retaining most of the tube output. Math and test gear/etc. GOOGLE is your friend.........
The plate current is set by the bias.  The plate voltage may vary a few volts as plate current increases or decreases.  The purpose of setting the bias is to find the optimum setting for controlling current in the output tubes or transformer.  Too much bias "hot bias" could damage the output tubes or transformer.  I was not aware of bias having any consideration whether an amp was ultralinear.  I always thought an amp with tapped output transformer for screen voltage was an ultralinear amp.

electra225

  • Hero Member
  • *****
  • Posts: 3221
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #10 on: January 14, 2019, 05:54:27 PM »
HFF, I had a senior moment and need to fall on my sword.   :-[


I got my "ohms per volt" backwards.   :-[


The specs for output from pin 7, the cathode, of the 6X4 rectifier is 300 volts.  This with the line voltage at 117 volts and measuring the voltage with a VTVM with 1000 ohms per volt sensitivity.  Your digital meter, along with 122 volt line voltage may be expected to yield higher voltage readings than those given in the manual.  In either case, I would not be terribly concerned about the higher readings you got.  And, again, if I was concerned, I would run the amp on a Variac, set to 117 volts line voltage, and take the readings with a VTVM.  There is a resistor, R58, 500 ohms at 7 watts that cuts the voltage from the rectifier to the output stage.  If that resistor has drifted, the readings might vary some as well.  What are the voltage readings on other elements in the amp?  If these were also high, line voltage would be a suspect for me.  There is probably some fancy mathematical formula for the relationship between input line voltage and the output in the secondary of the transformer, but that would be above my pay grade.  Good luck.


Sorry about hijacking your thread.   :-[
I don't need Google.  My wife says she knows everything.

ed from Baltimore

  • Sr. Member
  • ****
  • Posts: 380
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #11 on: January 15, 2019, 11:22:38 AM »
            The DC screen voltage is same as plate DC voltage in an ultra-linear power amp unless the screens connect to an entirely  separate screen which usually  fed by a lower voltage DC source as in the ACRO transformer made for 6146 output tubes. In that circuit the plate winding connects to 600 VDC, screens 300 VDC..
          The term ultra-linear, which was I believe first used in the Audio Engineering article by Herb Keroes and Dave Hafler, the partners  in ACRO products  in their patent application from 1952 (patent granted 1955)  refers to any power amplifier circuit in which a only a partial amount of the plate to plate load is shared by the screens. Percent refers to the portion of power, not voltage, fed to the screen. It has nothing to do with screen DC voltage. In their patent application the inventors mention different tubes having an optimum percentage power ratio, 18,5 % for 6L6, 5881 etc, only 5 % for 6V6. I remember reading in another article somewhere that 43 % was optimum for 6550 and that is the same number mentioned by VOX ac30 so the person he was quoting must have seen the same test. 
        Google Dave Hafler Herb Keros ultralinear patent, and ACRO transformer catalog for more info. Patent app has interesting graphs showing variations in di=storton with percent loading from  0 to
- 100

ed from Baltimore

  • Sr. Member
  • ****
  • Posts: 380
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #12 on: January 15, 2019, 12:09:05 PM »
           Old VTVMs had peak-to-peak rectifiers that convert AC to pulsating DC voltage with a capacitor to charge up and retain the highest peak between cycles. Some had 6AL5 or 6H6 dual diodes in a full wave circuit, some had a capacitor-diode series circuit with an RC filter following, but either circuit produced a DC voltage equal to the peak-peak AC. The calibrating pot in series with the panel meter that was connected in on the AC setting only was adjusted to give a meter reading that would be the same as a pure sine wave of the same RMS value. Usually the line voltage is a flattened looking waveform not a true sine wave, because so many circuits with rectifiers as in most power suppies only conduct current during the voltage peaks. The VTVM peak rectifiers in a circuit calibrated for correct reading on a pure sine wave would give a lower reading than a modern DMM which either has an average responding AC rectifier circuit or a true RMS DC to AC converter so they would indicate a higher reading than a VTVM looking at a flattened AC waveform.
           An increased line voltage puts more voltage on the 5 VAC filament winding on the rectifier tube which makes it a better rectifier. I mi=ght vary just the 5VAC winding one day to see how much it contributes to a higher DC output voltage. 

electra225

  • Hero Member
  • *****
  • Posts: 3221
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #13 on: January 15, 2019, 12:13:30 PM »
Thanks, Ed!  As usual, I learned something.  Good to see you back on the forums!   :) :) :) :)
I don't need Google.  My wife says she knows everything.

ed from Baltimore

  • Sr. Member
  • ****
  • Posts: 380
    • View Profile
Re: V-M 557 Plate/Screen Grid Voltage Questions
« Reply #14 on: January 15, 2019, 12:39:35 PM »
           I meant AC to DC converter. The computer at the Marriott I'm staying has a 30 minute reboot feature and I always hit POST REPLY at the last possible moment without a chance to proofread.
           I have noticed that DC voltages in old tube amplifiers are always much higher than schematic values, even when allowing for the increased line voltage. I have always wondered if they use a lower than stated voltage, say 110 instead of 117 VAC when generating the schematic just to avoid lots of feedback from customers worrying about getting "low" readings and thinking they were somehow "gypped". Like buying a bottle of soda and complaining if the fluid level looks low, but not if it's at the tip top of the neck.
       Old output tubes that are weak and not conducting as much as they should would cause the DC voltage to be higher as the current drawn from the circuit would be down and the filter caps would not be discharged as much between rectified pulses.  Of course an amplifier with adjustable bias set for a certain current wouldn't be affected but if the  bias pot  is just a balance pot to set the currents to equal values and not to a set value, it would be affected. If the voltage rating of the filter capacitor is a lot higher than the DC voltage measured, the capacitor eventually reforms to a  capacitor with a lower voltage rating but higher capacitance and the higher capacitance would discharge less between pulse peaks which would give a higher DC voltage.