Measuring Tube Gm - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 12th September 2009, 12:19 AM   #21
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Would high value (100K?) series limit resistors be effective to prevent the discharge, since they would limit the current?

One would have to take them into account in measurements with them in series with the meter.
  Reply With Quote
Old 12th September 2009, 02:12 PM   #22
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
I suspect something less than 100K might suffice. One of the potential concerns with a resistor and doing an ac based gm measurement except at very low frequencies might be the effect of the meter's input capacitance - you could compensate of course with the appropriately sized capacitor across your current limiting resistor. And yes you would need to compensate for the effect of the resistive voltage drop. The other option would be just to get a couple more inexpensive meters and not switch things at all.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 13th September 2009, 04:03 AM   #23
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
It may be easier to measure the cathode current separately from the rest of the measurements, since this is specified in the RCA data sheet and needs to be set. I'm setting the cathode current by adjusting the grid bias, then adjusting the plate and screen voltages.

I rebuilt the amp sections, and took readings on 13 tubes.

I measured the voltage drop across the cathode resistor (68 Ohm) to set the cathode current by adjusting the grid bias. Once it was set to .734V +/-0.5%, I adjusted the plate and screen voltages to 100V +/- 0.5%. 0.5% is about as good as I can get with single turn pots for adjustments.

I also measured the tubes on the Mercury 1000 for comparison.

The Gm varied from 2670 to 5333 on my test set, and 2000 to 4200 on the Mercury.

# Test Set V-Grid Mercury
1 2670 -1.34 2000
2 2700 -1.74 3950
3 2700 -1.66 3900
4 3060 -1.42 3950
5 3130 -1.75 3250
6 3250 -1.81 2600
7 3250 -1.41 3900
8 3250 -1.29 4100
9 3360 -1.25 4150
10 3360 -1.29 4350
11 3360 -1.16 4100
12 3480 -1.05 3350
13 5333 +0.10 4200

#s 5,6,12, and 13 are interesting as they don't follow the main trend.

5 and 6 show a decrease in grid voltage to achieve the same cathode current, with a drop in Gm on the Mercury test set.

12 shows a drop in gm on the mercury, with an increase on the test set.

13 shows a major jump in Gm on the test set, and the highest Gm on the Mercury. It also is the only tube requiring positive grid voltage to achieve 10.8mA of cathode current.

I presume the +21-44% difference between the test set and Mercury is due to the differences in the way they measure Gm, although I expected to see a better correlation.

These tubes are of unknown status as they came from a TV repair shop that didn't keep track of which tubes were used and which were new. I suspect most if not all are used.

I've purchased 10 NOS 5749W/6BA6Ws. Once they come in, I'll try to see if I can get more consistent results.
  Reply With Quote
Old 16th September 2009, 03:57 AM   #24
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
How dependent is Gm with respect to the AC screen drive voltage?

The definition simply states a change in plate current with respect to a change in grid voltage with the plate voltage held constant, and RCA RC-30 seems to imply that a 1V signal should be used (P103, Fig. 131).

I was testing with a 1V-rms input because it was easy to measure the change in output current, but I tried a 0.1V rms and now a .01Vrms input and I see a marked difference in readings. As I decrease the input voltage, I see an increase in Gm.

By the time I get to 0.01Vrms I am seeing a Gm in the 4000 range compared to 3200 at 1Vrms input and 3400 at .01Vrms.

This is more in line with what I expect from the tubes I have, and once I calibrated the Mercury 1000 based on these readings, the 12 KT88s a friend asked me to test all dropped the range I expected. They were reading 18% low before I re-calibrated the Mercury 1000 based on the latest test data on the 6BA6s.

Steven

Last edited by TheGimp; 16th September 2009 at 04:02 AM. Reason: Clarification, not talking about DC bias, but ac drive
  Reply With Quote
Old 16th September 2009, 05:43 PM   #25
diyAudio Member
 
HollowState's Avatar
 
Join Date: Aug 2006
Location: Taxland, New Jersey
Quote:
Originally Posted by TheGimp View Post
How dependent is Gm with respect to the AC screen drive voltage?
I hope this was just a late night typo. Dynamic Gm testing is traditionally done with a signal on the control grid (G1), not the screen. And a 1 volt signal level seems to be the accepted standard. Although I suppose a 0.1 volt level could be used as well, or for very sensitive small signal tubes.

As the input level goes down, the accurate measurment of shrinking AC plate currents can become more challenging. Effective shielding along with a good common ground between instruments is necessary. Sometimes noise and stray pickup will affect readings. I think you're using op-amps in your measurements somewhere. Small offset currents that weren't noticeable at higher levels may be affecting things at very low levels. Perhaps even thermal noise in the tubes may play a part at these low magnitudes.
__________________
"It is hard to imagine a more stupid or more dangerous way of making decisions than by putting those decisions in the hands of people who pay no price for being wrong." ~Thomas Sowell
  Reply With Quote
Old 16th September 2009, 08:00 PM   #26
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Yes, late nite working and fat thumbing it got screen instead of grid.

Yes, AD629 Instrumentation amp followed by two gain stages to measure the current through the sense resistor.

Since I'm interested in the 1KHz test signal and taking AC measurements, I would not expect offset to be an issue. I did null the op-amps to get less than 10mv offset just to be sure I was not driving any stage to clipping.

With 1Vrms Input, I was concerned that I might be pushing the tube out of linear operation. Although, 1Vrms and Gm of 4100 should put the excursions around +/-5.8mA. With a bias point of 10.8mA that should remain in the linear range.

I picked up a new meter today at Radio Shack (22-812). I'll compare the results with it to what I get with my other meter. Also this meter has a low scale of 400uA, so I might just put it in the cathode circuit and directly measure ac current with it to see if I get a better measurement that way.
  Reply With Quote
Old 17th September 2009, 02:56 AM   #27
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Well, I'm still getting readings that I believe are 20-25% low.

I believe the readings are low because I can use the measured tubes to calibrate the Mercury 1000 test set, and then when I test new tubes, I get readings that are consistently 20-25% below what I expect.

Therefore I must be doing something wrong in trying to measure Gm of the 6BA6s. The Screen and Plate supplies are regulated with 0B3s to around 108V, then followed by 5K pots with 220uF caps as final filters to set the actual screen and plate voltages.

I placed the new meter connections in the plate drive (cap is on the supply side of the meter, not at the plate). My procedure is :

1. Set filament voltage to 6.3Vac
2. adjust grid voltage to achieve a plate current of 10.8mA (new meter directly in plate circuit measuring DC Current).
3. Adjust screen and plate voltages to 100.0VDc
4. Re-Adjust screen voltage to get plate current back to 10.8mAdc if it has shifted
5. Verify screen and plate voltages haven't shifted, repeat 3-5 as necessary.
6. Verify 1KHz grid drive is 1.00V (this changes from tube to tube, so my drive impedance is probably too low, but can be adjusted)
7. Switch meter in plate circuit from DC to AC and measure plate current
8. since grid drive is 1vrms, plate current in mA rms represents Gm.

I've replaced all the electrolytic caps, and a few power resistors that were out of tolerance in the test set.

Steven
  Reply With Quote
Old 17th September 2009, 03:31 PM   #28
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
I believe my mistake was a misinterpretation of the data sheet.

The data sheet under "Characteristics" calls out two configurations. One at 100V Plate and one at 250V Plate. I'm using the 100V Plate specs which are :

Plate Voltage - 100V
Grid-No3 and internal shield - Connected to cathode at socket
Grid-No2 supply voltage - 100V
Cathode Bias Resistor - 68 Ohm
Plate Resistance, approximately - 0.25 Meg
Transconductance - 4300 umhos
Plate Current - 10.8 mA
Grid-No2 Current - 4.4 mA
Grid-No1 voltage (approx) for transconductance of 40 umhos - --20Volts

Since I am adjusting the Grid-No1 bias to set the plate current to 10.8mA, I should not have had a cathode resistor in the circuit. The cathode resistor was providing negative feedback which resulted in low measured Gm values.

Steven
  Reply With Quote
Old 17th September 2009, 04:32 PM   #29
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Having removed the offending cathode resistor you should be getting correlation within a few % or better of the published numbers which were usually based on the averaged measurements of a large number of samples.

Sounds like quite a worthwhile project and with what you are learning you will be able to determine the transconductance of just about anything out there under realistic operating conditions.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 18th September 2009, 04:22 AM   #30
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
I got lucky and my 10 JAN5749Ws arrived this morning. The seller actually sent me 12 tubes, which was fortunate since two had very low Gm.

I suspect the JAN5749W is a heavy duty version of the 6BA6, and has lower GM overall based on this sample (that or I got un-lucky).

I compared the Gm measured with my test set and the results from my Mercury and got the following (first number is my test set, second number is Mercury):
3270 - 4150 26.9%
3620 - 3600 -0.5%
3750 - 3600 -4.0%
3760 - 4250 -13.0%
3840 - 3700 3.6%
3850 - 3900 -1.29%
3870 - 3500 -8.2%
3800 - 3880 2.06%
3910 - 4200 7.42%
3910 - 4250 8.69%
3950 - 4000 1.27%
4020 - 3950 -1.74%

1, 4, 6, 7 (possibly 9) and 10 seem to have a lot of variation. I suspect this is due to the difference in test methods between standard Gm testing and the way the Mercury tests tubes (more generic).

I guess I could do an Excel spreadsheet and come up with a better cal point that would even out the spread, but it's too late at night for that.

In fact that might be a better way to calibrate the test set.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
wm-61 measuring mic nukaidee Multi-Way 55 25th March 2009 11:44 PM
how is it measuring? fumihiko Digital Source 18 24th January 2005 04:10 PM
Measuring Le thatvan Multi-Way 6 29th October 2004 10:09 PM
Measuring fT and hFE? capslock Solid State 6 23rd September 2003 03:50 PM
Easy ways of measuring output impedance of tube stages stigla Tubes / Valves 4 16th May 2003 11:37 AM


New To Site? Need Help?

All times are GMT. The time now is 07:04 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2