Sorry I don't have a slew of data to offer you guys yet, and making a schematic takes almost as long as breadboarding (lol), but...basically what I have is an AA864 PI and pwr section, two stages of preamp (100K loads, 1500 ohm cathode bias), and a 2-pot AA763 ConcertAmp tone stack. Had to play with B+ rail resistors cause I'm not using a choke, and the PT is the Weber W022798INT. Using the 660VAC sec windings. Bias winding is 45VAC. Full B+ from the diodes goes to the OT center tap (419VDC quiescent), 220R drops to screens (417V), 1200R drops to splitter (408V, plates get 270V), and 10K drops to preamp plates (368V, plates get 195V). I raised grid stopper to LTP from 220K to 1M due to oscillation (it's on a breadboard). Bias circuit same as AA864. -42.5V bias to the grids gives me 50mA cathode mA. Oh, tubes are 12AX7 (both pre and PI), and 6L6s are matched Sovtek 6L6WXT. I swap in a pair of 6V6GTs (Russian Tungsol) and almost don't have to rebias! Set at 55mA (cathode) for the 6L6s, and it changes to 20mA with the 6V6s... Also, I didn't hook up the NFB (to OT sec)...left it for last, not sure which OT lead to use, and the amp sounds so good I never looked back.... I WILL look at bias voltage with volume up, screen current (to get a better handle on true plate mA), Sine wave clipping to square wave (am I class A,AB,B?), using both scope channels to see power tube phase overlap and distortion, etc etc. Too much fun stuff, too little time. Thanks again for all the comments and information!
Attachments
I'm looking at the plates of the 6L6s (pin 3 to gnd) on my scope, want to see what Tone Lizard shows on his scoping a Princeton... Is he looking at the plate of one tube? I want to see that "crossover notch". My scope shot shows what looks like maybe slight oscillation? Not harmonics or noise....not sure. put both tubes on both scope channels, Then noticed (arrrg!) some red-plating....vol was about 7 (highest clean), but cathode mA was 70 (from a constant sig gen to a dummy load).... What my cell pic shows blue was actually orange (color balance off?). Just wondering what you guys have seen on your scopes from tube amps, and what I'm looking at.. Thanks!
Attachments
OK, got some data...measured cathode current and plate current (have an octal extender with a 1R 1% on pin 3). The following is mA-plate / mA-cathode, from 0 - 10 vol: 44.1/50.0, 53.6/55.0, 58.0/60.0, 62.6/65.0, 66.9/70.0, 93/100, 114/125, 116/127. I grabbed the last three sets quickly, not wanting the tubes to red-plate (they didn't). Seems plate current is not going much past max rating, and screen current (cathode-plate) goes from 2-11 mA. Scope shows pretty squared wave at max vol, but screen current doesn't seem to runaway... Only have 2 DMMs, but can measure one current and one voltage (plate then screen) with subsequent tests, to fill out a usable data sheet. More to come!
Arrrg!!?? I'm running my amp up, getting screen and plate voltages back-and-forth with one meter, at the same cathode current points from my first test, and see screen voltage starting from being higher than plate, but dropping below plate voltage towards full vol - interesting... so I hook both meters to measure both voltages (to find out where they "cross over"), and notice one meter reading change by over 10 volts JUST BY CONNECTING THE OTHER METER TO THE AMP.....Arrrg!! WTF? Do I need a VTVM? Don't Fluke meters have a high enough resistance? Grrr! Back to the drawing board......
Sorry if I'm confusing anyone.... all my recent posts revolve around a breadboarded Bassman amp which I want to be able to run either 6V6s or 6L6s in. I have succeeded in that goal. Using components that will handle a 6L6 (PT, OT, etc) in the amp, I can now simply swap tubes, with minimal bias adjustment (don't ask me how that happened, but it does). This thread relates to either tube used, although of course the data will be a lot different! I just happen to have had 6L6s in it when I got curious about quiescent and dynamic currents. I now know what red-plating looks like (LOL), but more important is when it happens and what causes it.
I wonder if you have succeeded. We still have not figured out the answer to your original question that 80mA per tube on 6V6s was OK and you at one point red plated your 6L6s. And when you got your 1-10 readings, you hurried to get through the last three so they wouldn't red plate. Am I missing something here?
Also, how have you handled the different OT primary impedance requirements of 6L6s and 6V6s? What is the primary impedance of your OT? You do sound like you have solved the bias differential problem.
Also, how have you handled the different OT primary impedance requirements of 6L6s and 6V6s? What is the primary impedance of your OT? You do sound like you have solved the bias differential problem.
Sorry! I've been jumping around too much with my testing..... I red plated my tubes because my bias pot was flakey, and would jump bias to full bias supply voltage, wasn't a runaway issue. Changed pot, but still wary.....and am still wary when I dime the amp, but my 6L6s handled 125 mA for a short time with amp cranked with constant Sig gen signal, no red plate. Backed it down after 20 sec or so. Good point about OT prim impedance, It has multiple sec taps (4 & 8 ohms), I will check tube data and see if I am mismatching. Can change dummy load to 8 and use 4 ohm tap, or vice versa.
Hi Guys
The original post questioned the importance and relevance of voltmeter readings across a cathode current-sense resistor at full power compared to idle. The idle reading is what the resistor is there to provide using safe ground-referenced voltage measurements. Any other reading under any other condition has no bearing on the idle setting.
You cannot use a specific current value for each tube type as that does not protect the tube from over-dissipation. You MUST know the supply voltage of the amp in question AND the power rating for the tube to be installed. Then using Ohm's Law you can calculate the maximum idle current for the tube.
Yes, you can run the tube at its full power rating if you wish - Hiwatt always biased that way.
Yes, you can idle the tube anywhere below full and have different tonalities and touch responsiveness. Want a mesa tone? Just bias cold for a lifeless nondynamic tone.
Only the plate voltage Va matters. Vs (screen voltage) does not matter at all with respect to setting idle current.
It is better to err on the side of conservatism and simply consider PLATE power. Some guys add screen dissipation in so they can nudge the total current up, but there is not benefit whatsoever to this and it erodes safety margins and tube life.
Tube ratings are continuous. Over the audio signal cycle the average heat from the plate should not exceed the rating even though instantaneous dissipation will be much higher and much lower.
When you cranked the amp and read the average current drawn by the tubes, of course it is much higher than at idle since this is a class-AB amplifier. Remember that as the tube pulls more current through itself that current goes through the OT primary creating a voltage drop across the winding - less voltage across the tube. At the peak current, voltage across the tube might be as little as 50V, depending on the tube and the circuit, so the tube heat at that instant is quite low. The TUT-series books explain all of this in detail.
Have fun
The original post questioned the importance and relevance of voltmeter readings across a cathode current-sense resistor at full power compared to idle. The idle reading is what the resistor is there to provide using safe ground-referenced voltage measurements. Any other reading under any other condition has no bearing on the idle setting.
You cannot use a specific current value for each tube type as that does not protect the tube from over-dissipation. You MUST know the supply voltage of the amp in question AND the power rating for the tube to be installed. Then using Ohm's Law you can calculate the maximum idle current for the tube.
Yes, you can run the tube at its full power rating if you wish - Hiwatt always biased that way.
Yes, you can idle the tube anywhere below full and have different tonalities and touch responsiveness. Want a mesa tone? Just bias cold for a lifeless nondynamic tone.
Only the plate voltage Va matters. Vs (screen voltage) does not matter at all with respect to setting idle current.
It is better to err on the side of conservatism and simply consider PLATE power. Some guys add screen dissipation in so they can nudge the total current up, but there is not benefit whatsoever to this and it erodes safety margins and tube life.
Tube ratings are continuous. Over the audio signal cycle the average heat from the plate should not exceed the rating even though instantaneous dissipation will be much higher and much lower.
When you cranked the amp and read the average current drawn by the tubes, of course it is much higher than at idle since this is a class-AB amplifier. Remember that as the tube pulls more current through itself that current goes through the OT primary creating a voltage drop across the winding - less voltage across the tube. At the peak current, voltage across the tube might be as little as 50V, depending on the tube and the circuit, so the tube heat at that instant is quite low. The TUT-series books explain all of this in detail.
Have fun
struth-great comments! I have learned a lot from you guys, and playing with my breadboarded Baseman....am in no rush to do a final build. I feel I've succeeded in my orignal goal, can swap 6L6s for 6V6s, the bias stays in range (but can still be adjusted), and the amp sounds great! I'm moving on to rebuilding an Epiphone Les Paul studio goth that I picked up for $100. Has set neck, but needs repainting and pickups.
I'm glad you have achieved your goal! Remember about power and volume db. When doubling power, you get a 3db increase. So let's assume you get 20 watts with 6V6s and 50 watts with 6L6s. You only gain 4db with that increase in power. People are constantly amazed when they have a 1/2 power switch and say it isn't muck quieter. Well it isn't. It is only 3db quieter. The 1/2 power switch is more useful for tone difference than loudness difference.(Usually that switch is a pentode/triode switch).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.