|
Tube DIY Asylum Do It Yourself (DIY) paradise for tube and SET project builders. |
|
|
|
Tube DIY Asylum Do It Yourself (DIY) paradise for tube and SET project builders. |
|
In Reply to: Re: Have you tried the WE connection on a CCS? posted by danlaudionut on April 27, 2005 at 20:41:29:
Yep, I have tried the WE connected bypass cap with the CCS.In the 47PP the migration path was resistor only followed by resistor bypassed with 40 uf GE polly in oil. Bypassing the resistor brought things into better focus and improved the imaging. Next the 40uf poly in oil was replaced with a 12uf GE paper in oil. The paper in oil was a good improvement in the low level detail department.
A little later after a jab from our good friend Allen Wright I added the CCS to the output stage. At this point I could compare the CCS only with the CCS bypassed by the 12uf paper in oil. Could not tell the difference between both setups and decided that if it sounded just as good without the cap, leave it out. One less cap is good in my book. The only down side to leaving the cap out is a slight loss of power.
The Tabor amp started out with the 40uf GE F97 poly in oil across the CCS. The amp was having some stange oscillations at the onset of clipping and the cap killed them. Later, after getting the rest of the circuit dialed with proper grid/screen stoppers and operating points the cap is not needed so out it came. There was a loss of ~1.5 watts output when the cap came out, down to 7.5 watts from 9.
A couple of weeks ago the Tabor was modified to increase the current in the input stage. The feedback resistors were reduced to 68K from 130K. The input CCS values were adjusted to compensate for the higher current from the feedback resistors.
At this time each 6AU6 is running 4.1ma where it used to be 2.1ma. Overall, the sensitivity wend down some but not as much as you would think. As the input stage current goes up the transconductance also goes up to help compensate for the lower value of feedback resistor.
The 150 volt VR tube was replace with a pair of 105 volt regulator tubes to bring the regulated supply up to 210 volts. This was nessasary to drive the screens harder to get the grid bias of the input tubes back up after increasing the bias current.
Changing the combo of feedback resistor value and input stage current was one of the biggest improvements so far. Where the performance of the amplifier before the input/feedback mod was the best I had heard so far the amp now has a vibrancy that is stunning. All the good things the amp did before just increased like 10% but the big improvement is dynamics. Transients are a whole new ball game now.
I also played with the input stage screen circuit a bunch last weekend. It was bugging me that I really did not have a good grip on what was going on around the input tube. If you looked at the voltage waveforms in the input circuit the screens were moving about 50% as much as the plates. I suspected that the screens moving was making the input tubes more linear. If you look at the transconductance curve of the 6AU6 it is quite curved at the operating point. As the grid is driven more positive the tube draws more plate current moving up the transconductance curve. As the plate current increases the screen current also increases. With the resistance in series with the screen the screen voltage will drop as the screen current increases. The dropping screen voltage reduces the transconductance helping to offset the increase caused by the plate current increasing.
On to the test bench. If the screens were bypassed to the cathode with caps like a true pentode circuit the measured distortion doubled! Starting with the values in the Tabor screen circuit I added more resistance to each screen. As the screen resistors increased the overall gain was reduced, the 3rd harmonic distortion went down but the second went up. Compared to pure pentode mode both second and third harmonics drop together as the resistance of the screens increase. Further increase we see the second stop dropping then start to increase again. With the 6AU6 the values that I had just pulled out of the air for the screen circuit were about as good as it gets. I did find that the series 2.2K resistor between the wiper of the balance pot and the 210 volt regulator needed to come out. It was worth a 10% reduction in distortion at 1 watt.
I also built an input circuit mockup using 6BN11 compactron dual pentode tubes. The 6BN11 liked to see 18K resistors between the screen and the balance pot for lowest distortion. The 6BN11 basically performs the same as the pair of 6AU6's except the sensitivity is higher, 1.4 volts RMS for full power where the 6AU6 needs 1.9 volts RMS for full power.
On the output stage I found that if the outputs were cranked up to 74ma each at 330 volts (24.4 watts disapation) the output power went up to 11 watts RMS at the onset of clipping, ~3% distortion. I think I like the sound better at the lower operating point of 68ma and 290 volts. Not much of a difference but I don't need the power (or heat) so I'll leave the amp at the lower op point for now.
Thats enough rambling for now...
Gary
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Have you tried the WE connection on a CCS? - Gary P 04/27/0521:50:06 04/27/05 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
