F6 Amplifier
 User Name Stay logged in? Password
 Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read Search

 Pass Labs This forum is dedicated to Pass Labs discussion.

 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
kasey197
diyAudio Member

Join Date: Sep 2009
hi all,

Had time to fire up the iron again this weekend and wanted to run some tests in open loop (ie no NFB/degeneration) to see if i understood things correctly. Here are some observations, theory and then tests for some proof.

Setup:

Same schema as papas front end (just the tx+output stage). Supply rails +\- 24V. Output devices 2sk1058 hitachi latefts. Bias arrangement: 9V batteries with 5K pot to set voltage. The battery and pot bypassed by 220uF caps. Separate bias for top and bottom halves. Bias current set for 1.4A. Transformer was regular torodial power transformer. Stepup ratio approx 4x.

Results:

0.2V pk-pk in, 9.8V pk-pk out. Therefore gain = 49x. Given the tx gives us 4x, that means the fets are giving us approx 12.2X gain.

Theory:

Load R = 6.8R, therefore combined bottom n top half transconductance = 12.2/6.8 = 1.8 Siemens. Each fet therefore gives us 1.8/2 = 0.9 S transconductance. This seems to tie in with what i see from the hitachi datasheets.

Explanation:
Given the above, both top and bottom fets are acting in common source mode (despite the temptation to think about the top half as a source follower providing no voltage gain). If this hypthesis is true then both top and bottom halves provide equal amounts of gain. We can test this easily given the transformer coupling by just bypassing the top and bottom tx secondaries (ie in turn, each fet would see only the bias voltage, but no signal) individually and seeing the effect.

Results:

Both secondaries connected, trace 1. Output is 9.8V.
Top secondary bypassed, trace 2,. Output is 4.875V
Bottom secondary bypassed, trace 3. Output is 4.812V

Summary: So, the sum of the two halves individually = 9.7V vs 9.8V when theyre both connected. Seems close enough to suggest that the initial hypothesis that both halves act to provide equal amounts of gain (in common source) is correct.

Some other results/thoughts:

Warts n all open loop, the distortion performance seems pretty remarkable by comparison to a conventional complmenetary push-pull stage built. Here, (trace 4) you can see that theres zilch even harmonics and all were left with is 0.07% THD, almost all of it 3rd harmonic with very little higher order stuff.

The op stage needs some local feedback so as to reduce the miller effect on input capacitance which is very high. Having said that, the open loop freq response incl tx was quite usefully flat (trace 5). Seems to me that this would run quite respectably as-is w/o nfb if one could live with the HF rolloff and high o/p impedance. Addition of some feedback on this relatively linear stage will make for a pretty nice amp i think.....

Next steps while waiting (for the real F6):

a) Try the same local biasing arrangement i did last time but with lower impedance resistors (need to grab some WW low R pots tom)

b) Build the dual jfet input stage to see how that works....

I'm finding that part of the fun is in the journey .....
Attached Images
 trace1.jpg (125.7 KB, 603 views) trace2.jpg (123.9 KB, 594 views) trace3.jpg (122.4 KB, 595 views) Trace4-THD.jpg (126.2 KB, 577 views) Openloopfreqresponse.jpg (133.3 KB, 573 views)

 5th August 2012, 05:57 PM #562 kasey197 The Dastardly Dad of Three diyAudio Member   Join Date: Sep 2009 Also while we wait, anyone on this list sells the semisouth R100s ? Thought id order them in ahead...I usually get stuff from zhoufang but he seems to be away these last few weeks .....
Antoinel
diyAudio Member

Join Date: Jan 2012
Location: Pennsylvania
Quote:
 Originally Posted by kasey197 hi all, Had time to fire up the iron again this weekend and wanted to run some tests in open loop (ie no NFB/degeneration) to see if i understood things correctly. Here are some observations, theory and then tests for some proof. Setup: Same schema as papas front end (just the tx+output stage). Supply rails +\- 24V. Output devices 2sk1058 hitachi latefts. Bias arrangement: 9V batteries with 5K pot to set voltage. The battery and pot bypassed by 220uF caps. Separate bias for top and bottom halves. Bias current set for 1.4A. Transformer was regular torodial power transformer. Stepup ratio approx 4x. Results: 0.2V pk-pk in, 9.8V pk-pk out. Therefore gain = 49x. Given the tx gives us 4x, that means the fets are giving us approx 12.2X gain. Theory: Load R = 6.8R, therefore combined bottom n top half transconductance = 12.2/6.8 = 1.8 Siemens. Each fet therefore gives us 1.8/2 = 0.9 S transconductance. This seems to tie in with what i see from the hitachi datasheets. Explanation: Given the above, both top and bottom fets are acting in common source mode (despite the temptation to think about the top half as a source follower providing no voltage gain). If this hypthesis is true then both top and bottom halves provide equal amounts of gain. We can test this easily given the transformer coupling by just bypassing the top and bottom tx secondaries (ie in turn, each fet would see only the bias voltage, but no signal) individually and seeing the effect. Results: Both secondaries connected, trace 1. Output is 9.8V. Top secondary bypassed, trace 2,. Output is 4.875V Bottom secondary bypassed, trace 3. Output is 4.812V Summary: So, the sum of the two halves individually = 9.7V vs 9.8V when theyre both connected. Seems close enough to suggest that the initial hypothesis that both halves act to provide equal amounts of gain (in common source) is correct. Some other results/thoughts: Warts n all open loop, the distortion performance seems pretty remarkable by comparison to a conventional complmenetary push-pull stage built. Here, (trace 4) you can see that theres zilch even harmonics and all were left with is 0.07% THD, almost all of it 3rd harmonic with very little higher order stuff. The op stage needs some local feedback so as to reduce the miller effect on input capacitance which is very high. Having said that, the open loop freq response incl tx was quite usefully flat (trace 5). Seems to me that this would run quite respectably as-is w/o nfb if one could live with the HF rolloff and high o/p impedance. Addition of some feedback on this relatively linear stage will make for a pretty nice amp i think..... Next steps while waiting (for the real F6): a) Try the same local biasing arrangement i did last time but with lower impedance resistors (need to grab some WW low R pots tom) b) Build the dual jfet input stage to see how that works.... I'm finding that part of the fun is in the journey .....
Great work kasey. You've separated [and delivered] fact from fiction.

lhquam
diyAudio Member

Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
Quote:
 Originally Posted by kasey197 hi all, Had time to fire up the iron again this weekend and wanted to run some tests in open loop (ie no NFB/degeneration) to see if i understood things correctly. Here are some observations, theory and then tests for some proof. ... I'm finding that part of the fun is in the journey .....
Excellent. How are you driving the transformer? The suggested Jensen transformer is JT-123-FLPCH which has 4 windings with equal numbers of turns. Thus it must be run either 1:1 or 2:1. I suspect that it should be run 1:1 with the 2 primary windings paralleled, which will reduce the gain by a factor of 4 vs. your transformer configuration. That will not leave a lot for negative feedback. It also suggests that the degeneration resistors should be kept low in order to keep the transconductance as high as possible, but having the MOSFETS be thermally stable.

kasey197
diyAudio Member

Join Date: Sep 2009
Quote:
 Originally Posted by lhquam Excellent. How are you driving the transformer? The suggested Jensen transformer is JT-123-FLPCH which has 4 windings with equal numbers of turns. Thus it must be run either 1:1 or 2:1. I suspect that it should be run 1:1 with the 2 primary windings paralleled, which will reduce the gain by a factor of 4 vs. your transformer configuration. That will not leave a lot for negative feedback. It also suggests that the degeneration resistors should be kept low in order to keep the transconductance as high as possible, but having the MOSFETS be thermally stable.
You have a fair point with the fets im using since they have low transconductance.... The r100s have more than 10 times as much transconductance so there should be PLENTY of gain end to end even with 2:1 step down tx (!) or the 1:1 ...

Re your first question, the test setup is driven direct by the hp 35665a signal analyzer which has a very low impedance..... Real world use is going to need the fet buffers ala f6 schematic. That's on the to-do list (eventually) after I try the local biasing n fb network

Tea-Bag
not politcally affiliated
diyAudio Member

Join Date: Feb 2006
Location: Kennebunk
Quote:
 Originally Posted by kasey197 Also while we wait, anyone on this list sells the semisouth R100s ? Thought id order them in ahead...I usually get stuff from zhoufang but he seems to be away these last few weeks .....
I sell the JFETs for less than Newark does. PM for details.
__________________
GB 2017 #2 Ends 12/3/2017 Click Here to Sign Up! | Hot-Rod DCB1 boards and kits In Stock, PM | M2C Boards In Stock

 5th August 2012, 08:14 PM #567 buzzforb   diyAudio Member     Join Date: Oct 2010 Location: Burlington, NC TeaBag is excellent to deal with. Much more prompt than I. __________________ ...Shape the sound , Man!
 5th August 2012, 08:25 PM #568 generg   diyAudio Member     Join Date: Jul 2009 Location: Near Frankfurt or me..... thank you kasey197! __________________ and the First Wtt is......a really First Class PSS Idea
 5th August 2012, 08:54 PM #569 Zen Mod   Official Court Jester diyAudio Member     Join Date: Jan 2003 Location: ancient Batsch , behind Iron Curtain __________________ my Papa is smarter than your Nelson ! clean thread; Cook Book;PSM LS Cook Book;Baby DiyA ;Mighty ZM's Bloggg;Papatreasure;Papa...© by Mighty ZM
Zen Mod
Official Court Jester
diyAudio Member

Join Date: Jan 2003
Location: ancient Batsch , behind Iron Curtain
Quote:
 Originally Posted by kasey197 ........ Setup: Same schema as papas front end (just the tx+output stage). Supply rails +\- 24V. Output devices 2sk1058 hitachi latefts. Bias arrangement: 9V batteries with 5K pot to set voltage. The battery and pot bypassed by 220uF caps. Separate bias for top and bottom halves. Bias current set for 1.4A. Transformer was regular torodial power transformer. Stepup ratio approx 4x. ....
had some C in series with secondaries , to save them of DC ?

Quote:
 Originally Posted by kasey197 ...... Seems close enough to suggest that the initial hypothesis that both halves act to provide equal amounts of gain (in common source) is correct. ......
yup

as I wrote - quasicomplementary is in function same as complementary ; with even greater symmetry between windings (which we can expect from signal xformer , comparing to crude mains one ) and , say , if we can use really dynamically matched (even best in situ , observing THD spectra ) outputs , we can expect even greater amount of suppression of even harmonics ; say that with some feedback that will be less critical than in case of OL .

Quote:
 Originally Posted by kasey197 ..... a) Try the same local biasing arrangement i did last time but with lower impedance resistors (need to grab some WW low R pots tom) ......
having more than , say , 3-5mA through bias net is hardly of any benefit ; remember that you need to decouple gate from bias net with at least 10K , to avoid unnecessary loading of secondaries

then secondary's action will give enough AC mA for proper feed of gates

that's what counts , not brute force in bias net

anyway - did you observe near clipping action of tried OS ?
__________________
my Papa is smarter than your Nelson !
clean thread; Cook Book;PSM LS Cook Book;Baby DiyA ;Mighty ZM's Bloggg;Papatreasure;Papa...© by Mighty ZM

 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 Forum Rules
 Forum Jump User Control Panel Private Messages Subscriptions Who's Online Search Forums Forums Home Site     Site Announcements     Forum Problems Amplifiers     Solid State     Pass Labs     Tubes / Valves     Chip Amps     Class D     Power Supplies     Headphone Systems Source & Line     Analogue Source     Analog Line Level     Digital Source     Digital Line Level     PC Based Loudspeakers     Multi-Way     Full Range     Subwoofers     Planars & Exotics Live Sound     PA Systems     Instruments and Amps Design & Build     Parts     Equipment & Tools     Construction Tips     Software Tools General Interest     Car Audio     diyAudio.com Articles     Music     Everything Else Member Areas     Introductions     The Lounge     Clubs & Events     In Memoriam The Moving Image Commercial Sector     Swap Meet     Group Buys     The diyAudio Store     Vendor Forums         Vendor's Bazaar         Sonic Craft         Apex Jr         Audio Sector         Acoustic Fun         Chipamp         DIY HiFi Supply         Elekit         Elektor         Mains Cables R Us         Parts Connexion         Planet 10 hifi         Quanghao Audio Design         Siliconray Online Electronics Store         Tubelab     Manufacturers         AKSA         Audio Poutine         Musicaltech         Aussie Amplifiers         CSS         exaDevices         Feastrex         GedLee         Head 'n' HiFi - Walter         Heatsink USA         miniDSP         SITO Audio         Twin Audio         Twisted Pear         Wild Burro Audio

 Similar Threads Thread Thread Starter Forum Replies Last Post Nounours18200 Pass Labs 37 31st August 2012 05:35 PM ioro Multi-Way 4 15th April 2012 09:27 PM jugoslavija Pass Labs 1 1st November 2011 08:39 AM kkosiba Multi-Way 14 9th November 2007 09:38 AM traw Pass Labs 7 2nd November 2007 08:09 PM

 New To Site? Need Help?

All times are GMT. The time now is 06:52 PM.