Sorry I thought answer this before. Things have been a bit askilter lately. My dad had a stroke last week and I just got back home from staying with him since last Wednesday.
I don't see what I did with the GNFB, but it was a very low level.
I went through the schematic and verified everything but the GNFB which is not attached to the prototype so I don't know what values I used. Here is what I believe I have currently on the prototype.
I don't see what I did with the GNFB, but it was a very low level.
I went through the schematic and verified everything but the GNFB which is not attached to the prototype so I don't know what values I used. Here is what I believe I have currently on the prototype.
An externally hosted image should be here but it was not working when we last tested it.
I've started working on the chassis and have the plate cut to 11" square. In order to drill the tube socket holes I decided to make a jig to allowe me to replicate the hole pattern more easily than laying it out with pen on masking tape. I only have to mark and drill a 1/4" center hole, mount the jig, and drill each pilot hole with the steel guide in place. Then drill out the holes to the proper diameter and punch the center hole. 5 Minutes of work per tube.
Here is the sample I made to test the jig. The finished test shown is for a B9A. The B7G looked like it would not support ten vent holes unless they were far from the tube or small. I may redo both jigs later if I think I can improve them.
Here is a link to the thread Istarted in tools on how it works. I'll be adding more pictures there and answering questions if anyone has any. :
http://www.diyaudio.com/forums/equipment-tools/160909-drilling-jig.html
Here is the sample I made to test the jig. The finished test shown is for a B9A. The B7G looked like it would not support ten vent holes unless they were far from the tube or small. I may redo both jigs later if I think I can improve them.
An externally hosted image should be here but it was not working when we last tested it.
Here is a link to the thread Istarted in tools on how it works. I'll be adding more pictures there and answering questions if anyone has any. :
http://www.diyaudio.com/forums/equipment-tools/160909-drilling-jig.html
I just got back from taking care of pop for five days. I had plenty of time while setting with him in the hospital to try working on a load line for the 6P1P in pp with a Class AB1 load line and the Class A load line at 36mA.
This is in parallel with the Meng Yue Mini thread:
http://www.diyaudio.com/forums/tubes-valves/157134-meng-yue-mini-schematic-21.html
...but I don't want to confuse it so I'm putting the Load Lines here.
I’m trying to calculate the maximum power out if I push the 6P1P-PP to class AB1 by upping the B+ to 300V and dropping the bias point to 36mA.
Paraphrased from Turner Audio page:
loadmatch4-pp-beamtetrodes
B+ = 300V
Ia = 36mA
RL= 10K p-p, 5K single Plate class A, 2K5 Class B
Class A Load Line:
Establish the desired working point ‘Q’ (36mA at 300V)
Plot Class A Load Line through 36mA bias at 300Vwith a slope of 5K. Current = 300V/5K = 60mA.
60mA + 36mA = 96mA
36mA*5K = 180V. 180V + 300V = 480V. Line is from 96mA at 0V to 480V @0mA. (Red Line)
Class B load line:
Drop a vertical line from the Bias Point Q to the Ea Axis and label it ‘D’.
Plot Class ‘B’ load line from B+=300V (‘D’) to Imax = 300V/2K5=120mA (point ‘A’).
Draw the Class B Load Line from ‘A’ to ‘D’ (Blue line).
Calculate Output Power:
Class A:
Drop a line from the point where the 5K load line crosses the 0mA Grid line (‘B’ = 55V)
Drop a line from the point where the two load lines cross (‘C’ = 112V)
Calculate class A power for one tube, Power = (Ea at Q – Ea at C) Squared/2X RL =(300-112)^2/2x5000 = 3.53W
Calculate total class A power = 2X power for one tube = 7.06W
Class AB?
Calculate the maximum peak anode voltage swing = 2 X pk Swing at each anode x 0.707 Vrms.
Peak swing = (300-55) X 2 X 0.707 = 346.43Vrms
Calculate maximum class AB power =max Vrms squared / RL a-a =346*346/10000 = 12W
It looks to me like from a static dissipation standpoint this is biasing the tubes close to the same as 250V at about 48mA.
The orange line was an attempt at pushing the B+ to 340V, but it does not look like there is a lot of power to be gained with that.
Can one of the more experienced members confirm or deny if I have done the calculations properly?
Thanks.
This is in parallel with the Meng Yue Mini thread:
http://www.diyaudio.com/forums/tubes-valves/157134-meng-yue-mini-schematic-21.html
...but I don't want to confuse it so I'm putting the Load Lines here.
I’m trying to calculate the maximum power out if I push the 6P1P-PP to class AB1 by upping the B+ to 300V and dropping the bias point to 36mA.
Paraphrased from Turner Audio page:
loadmatch4-pp-beamtetrodes
B+ = 300V
Ia = 36mA
RL= 10K p-p, 5K single Plate class A, 2K5 Class B
Class A Load Line:
Establish the desired working point ‘Q’ (36mA at 300V)
Plot Class A Load Line through 36mA bias at 300Vwith a slope of 5K. Current = 300V/5K = 60mA.
60mA + 36mA = 96mA
36mA*5K = 180V. 180V + 300V = 480V. Line is from 96mA at 0V to 480V @0mA. (Red Line)
Class B load line:
Drop a vertical line from the Bias Point Q to the Ea Axis and label it ‘D’.
Plot Class ‘B’ load line from B+=300V (‘D’) to Imax = 300V/2K5=120mA (point ‘A’).
Draw the Class B Load Line from ‘A’ to ‘D’ (Blue line).
Calculate Output Power:
Class A:
Drop a line from the point where the 5K load line crosses the 0mA Grid line (‘B’ = 55V)
Drop a line from the point where the two load lines cross (‘C’ = 112V)
Calculate class A power for one tube, Power = (Ea at Q – Ea at C) Squared/2X RL =(300-112)^2/2x5000 = 3.53W
Calculate total class A power = 2X power for one tube = 7.06W
Class AB?
Calculate the maximum peak anode voltage swing = 2 X pk Swing at each anode x 0.707 Vrms.
Peak swing = (300-55) X 2 X 0.707 = 346.43Vrms
Calculate maximum class AB power =max Vrms squared / RL a-a =346*346/10000 = 12W
An externally hosted image should be here but it was not working when we last tested it.
It looks to me like from a static dissipation standpoint this is biasing the tubes close to the same as 250V at about 48mA.
The orange line was an attempt at pushing the B+ to 340V, but it does not look like there is a lot of power to be gained with that.
Can one of the more experienced members confirm or deny if I have done the calculations properly?
Thanks.
I can't confirm or deny and I'm not an experienced member, but the loadlines look close enough. I think that graph is for G2 = 250V so it will all go out the window if you use 300V on G2. I tried around 280V from anode to cathode on some 6P1P-EV for about 10 minutes and saw no red plating, they would have had 280V on the screens too. Not sure how robust the screen grids are on the -EV tubes, they have less max current in the specs compared to a 6P1P. I ordered some parts tonight to make 4 simple screen grid voltage regulators using LR8N3's, the experiment is to see what does it sound like if all 4 grid voltages are adjusted to the same voltage, and adjust each tube's bias current by adjusting the screen voltage, this might match up the tubes nicely, be interesting to hear if it sounds any better or not. Good luck with your directions too.
Ian.
Ian.
To make it ven simpler for you I have made an Excel-spreadsheet using Turners formulas:
http://www.revintage.se/PPABAMP.xls
And while you´re at it also try this one:
http://www.revintage.se/triodecalc.xls
This one shows distortion, max level, bandwith etc.
http://www.revintage.se/PPABAMP.xls
And while you´re at it also try this one:
http://www.revintage.se/triodecalc.xls
This one shows distortion, max level, bandwith etc.
Last edited:
Thanks Lars, I'll try your spreadsheet later. I made a simple one that only uses the values from the class AB calculation from LOAD MATCHING 4.
My calculations indicated that the lower 8Kp-p transformer should give more power out so I tried dropping the load resistor from 8Ohms to 6.4 Ohms to simulate a 8Kp-p transformer with a 8 ohm load and the output is now 7.8Vrms for 9.5W out with 36mA bias and 250V anode to cathode.
It looks like the lower p-p impedance gives higher power out, but I need to measure distortion to know how it is really going to behave as the output vs distortion curve looks like it will be worse.
I'll try upping the resistance to 9.6Ohms to simulate a 11Kp-p transformer to see if it follows the calculations I did on it's load line (should be lower power although the 6P1P-EV spec sheet shows it as the peak).
My calculations indicated that the lower 8Kp-p transformer should give more power out so I tried dropping the load resistor from 8Ohms to 6.4 Ohms to simulate a 8Kp-p transformer with a 8 ohm load and the output is now 7.8Vrms for 9.5W out with 36mA bias and 250V anode to cathode.
It looks like the lower p-p impedance gives higher power out, but I need to measure distortion to know how it is really going to behave as the output vs distortion curve looks like it will be worse.
I'll try upping the resistance to 9.6Ohms to simulate a 11Kp-p transformer to see if it follows the calculations I did on it's load line (should be lower power although the 6P1P-EV spec sheet shows it as the peak).
I've started construction
Time to quit playing and build the first amp.
I've got two sets of 10W transformers (salvaged Olson far left {10K?} and Hammond 10K on plate left corner) and one set of 30W EDCORs (5K on plate rt corner). The EDCORs will be used for a parallel push pull later on. I put them on the 11" Square plate to see how it would look and if it would fit or I need a bigger plate for it. It will get real crowded if I add two more output tubes per channel, but might work.
I'll get this one built first before I try anything else.
I've already tested the power supplies on the bench with the current prototype. It will be point to point wiring.
Time to quit playing and build the first amp.
I've got two sets of 10W transformers (salvaged Olson far left {10K?} and Hammond 10K on plate left corner) and one set of 30W EDCORs (5K on plate rt corner). The EDCORs will be used for a parallel push pull later on. I put them on the 11" Square plate to see how it would look and if it would fit or I need a bigger plate for it. It will get real crowded if I add two more output tubes per channel, but might work.
An externally hosted image should be here but it was not working when we last tested it.
I'll get this one built first before I try anything else.
I've already tested the power supplies on the bench with the current prototype. It will be point to point wiring.
Layout
Drill Pattern.
Right Channel Layout seen from the bottom.
Left Channel Layout seen from the bottom.
Next I have to figure out the AC in, fuse, and power switch (including how to mount them). Then the line inputs, and speaker outputs. Finally I need to figure out how to route the filament power, and where to put the single point ground.
Drill Pattern.
An externally hosted image should be here but it was not working when we last tested it.
Right Channel Layout seen from the bottom.
An externally hosted image should be here but it was not working when we last tested it.
Left Channel Layout seen from the bottom.
An externally hosted image should be here but it was not working when we last tested it.
Next I have to figure out the AC in, fuse, and power switch (including how to mount them). Then the line inputs, and speaker outputs. Finally I need to figure out how to route the filament power, and where to put the single point ground.
I got the panel drilled and everything test fitted:
Everything fit fine so far, but I need to get some tag strips before I can start wiring it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Everything fit fine so far, but I need to get some tag strips before I can start wiring it.
Well, I trashed 5 6P1P-EVs last night. I was checking tubes for Ia and Gm with a 510 Ohm cathode bias resistor in place when I picked a tube that went blue flash and dumped over 250mA out of the power supply. This tripped the 1/4A breaker, but it was obvious it was a bad tube. The next four tubes I unwrapped and tried all started around 30mA Ia and gradually crept up towards 100mA with a steadly increasing blue glow.
How could I get five bad tubes in a row after testing over 30 without a problem? I quit testing at this point and plugged one of the previously tested tubes in and as soon as I saw it start to glow blue I shut it down!
I stripped all the reisitors from the PP output stage and replaced them. Then I tried two previously tested good tubes and they seemed to work fine. The five tubes that tested bad still draw excessive current so I'll cut them up to look at the construction.
I finished testing a total of 50 tubes, with 5 bad.
The statistical probablility of finding that many bad tubes in a row has got to be very high. So I presume the first tube was bad and caused some kind of dammage to the circuit which induced the fault in the next four tubes.
The Screen resistors (1W carbon comp 1K) had shifted to over 2K each.
The cathode resistors (510Ohm CC) had shifted values and one measured in the Meg range, but I suspect it was arcing internally.
One anode (1 Ohm MF 1W) measured open, but again I suspect it was arcing internally or the tube couldn't have been running at all.
Would excessive current cause the blue glow? I thought a gassy tube would do that, but I was not aware of any way to induce it.
How could I get five bad tubes in a row after testing over 30 without a problem? I quit testing at this point and plugged one of the previously tested tubes in and as soon as I saw it start to glow blue I shut it down!
I stripped all the reisitors from the PP output stage and replaced them. Then I tried two previously tested good tubes and they seemed to work fine. The five tubes that tested bad still draw excessive current so I'll cut them up to look at the construction.
I finished testing a total of 50 tubes, with 5 bad.
The statistical probablility of finding that many bad tubes in a row has got to be very high. So I presume the first tube was bad and caused some kind of dammage to the circuit which induced the fault in the next four tubes.
The Screen resistors (1W carbon comp 1K) had shifted to over 2K each.
The cathode resistors (510Ohm CC) had shifted values and one measured in the Meg range, but I suspect it was arcing internally.
One anode (1 Ohm MF 1W) measured open, but again I suspect it was arcing internally or the tube couldn't have been running at all.
Would excessive current cause the blue glow? I thought a gassy tube would do that, but I was not aware of any way to induce it.
An externally hosted image should be here but it was not working when we last tested it.
The tube on the right is the worst one. The tube on the left can be seen to be throwing blue light, but not the flash of the one on the right. This combo pops a 0.5A circuit breaker in the primary so It is pulling at least 1/2A in the secondary.
I'll open them up to see if I can find anything obviously wrong.
I guess you were unlucky, because a 560 Rk should have limited the current through the tube quite dramatically.
The only thing I can think of is that the circuit started to oscillate, but again I would have thought that unlikely given it's a resistive test circuit. Did you have a grid stopper resistor?
The only thing I can think of is that the circuit started to oscillate, but again I would have thought that unlikely given it's a resistive test circuit. Did you have a grid stopper resistor?