Testing on latest PP 300B build - 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 5th August 2015, 02:11 PM   #1
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Default Testing on latest PP 300B build

Just finished the build/re-build of my 300B push pull amplifier. Schematic attached. I was formerly using a very close build of Lynn Olson's Karna amp, using two interstage transformers. I had challenges getting the input stage right, due to the difficulty in the driving of the IT. I have since performed some testing of different topologies, and settled on the design here. It was the optimal way I found of getting lowest distortion to the driver stage.

Also, I modified the driver stage to fully differential, as it also provided excellent performance at 15W and below drive levels. Even at higher drive where some 300B grid current started to flow, I still had pretty good performance of the driver stage. Changing out my IT (I had tried a few brands) to the bifilar Monolith IT-02 gapped for push pull provided a few dB of lower odd distortion as well.

Output stage has pretty much remained as is, and has now become the limiting factor in the design. Not that I feel any need to improve it, but it's nice that I have given the output stage everything possible to provide low distortion.

Finally, I have resorted to using Rod Coleman's regulators in both DHT stages. They meet up to the hype; performance is repeatable; the DC voltage is rock steady, and "zero-point-zero" hum as a result. More commentary to follow in the next few posts: it will be better organized to separate data in individual posts.

Basic build description is a single chassis for power supply, and mono block amps for signal path. DHT heating power supplies are incorporated into the mono blocks, as the initial build provided only a single 120VAC circuit for AC heating transformers. Happy to report that having the rectifier transformers in the mono blocks has been successful.

Everything is in the garage, so pictures are so-so, but you get an idea of the build.

I am interested in some commentary on the performance, as well as some assistance in taking meaningful measurements. I have decent hardware for measuring, but running Audiotester software has its challenges.
Attached Images
File Type: jpg IMG_0681.JPG (128.8 KB, 486 views)
File Type: jpg IMG_0705.JPG (93.1 KB, 472 views)
File Type: jpg IMG_0711.JPG (129.3 KB, 465 views)
File Type: jpg IMG_0712.JPG (91.9 KB, 468 views)
Attached Files
File Type: pdf schematic.pdf (15.8 KB, 211 views)
  Reply With Quote
Old 5th August 2015, 02:20 PM   #2
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
First, a shot of the output terminals with zero signal input. Very quiet, pleased to report. The 60 and 180 Hz shown here is actually a result of the rectifier transformers' magnetic field spilling into my measurement system. With shorted leads, I can actually produce this signal depending on the proximity to the transformers. It actually has little effect on the amp proper. Being in a garage with one bench, my test setup could use some improvement, but I have no issues with initial results like this. I have also attached my earlier design using AC heating, so you can see the improvement in noise at the output terminals. Three cheers for Rod's supply!
Attached Images
File Type: jpg idle.jpg (129.3 KB, 451 views)
File Type: jpg Left noise floor.jpg (164.4 KB, 123 views)
  Reply With Quote
Old 5th August 2015, 02:26 PM   #3
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Frequency response and distortion. I know my distortion graph needs some help; haven't figured that out yet. Help is appreciated.

This amp will be driving speakers actively crossed over at the 80-120 Hz range, so I'm not too concerned about the low frequency behavior, but generally I'm pleased. -1 dB within 20 Hz - 30 kHz. Not too shabby for a no feedback design.
Attached Images
File Type: jpg FR.jpg (125.3 KB, 142 views)
File Type: jpg distortion.jpg (106.4 KB, 125 views)
  Reply With Quote
Old 5th August 2015, 02:39 PM   #4
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Performance at 1W into 4 ohm resistive load. The captions near the top of the image explains the conditions.

First is the speaker output. 0.02%. Note the sidebands around the fundamental are 120 Hz, which originates from the 250mV ripple present on the output stage B+. The input and driver stage have no ripple effect, as the input stage benefits from the CCS rejection, and the driver stage is shunt regulated with gas tubes.

Next is the same conditions, but measurement is differential at the grids of the 46 triodes. This shows the input stage is doing a very fine job at 0.011%. Depending on the tube I use, I can improve the input stage performance. I have a few stellar SN7's that have been verified, and will use those when actually listening to the system. But for testing purposes I picked a good (not stellar) one.

Finally is the differential measurement at the grids of the 300B's. A little additional distortion has been added in the driver stage, but very minimal. I may experiment with matching pairs based on FFT and bias to try and improve the driver, but generally the triode connected 46 does a fine job.
Attached Images
File Type: jpg 1W.jpg (118.4 KB, 123 views)
File Type: jpg 1W-46 grid.jpg (116.0 KB, 90 views)
File Type: jpg 1W-300b grid.jpg (120.2 KB, 76 views)
  Reply With Quote
Old 5th August 2015, 02:46 PM   #5
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Performance at 15W into 4 ohm load. This will likely be my practical limit of output power, given my speakers and the volume level I listen to. Don't need rock concert levels, but I appreciate the power a few transients may require.

Again, speaker terminals, 46 grids, and 300B grids.

Input stage continues to operate very linearly, with friendly FFT spectrum and nonexistent higher order harmonics. Driver stage begins to add some higher order harmonics, but they are really really low given the signal drive level. Overall the result at the speaker terminals is pleasing; decent waterfall rolloff with low higher orders. Output stage seems to be holding its own at this point. The sidebands are annoying on paper, but I'm not about to regulate the B+, so it is what it is.
Attached Images
File Type: jpg 15W.jpg (117.3 KB, 87 views)
File Type: jpg 15W-46 grid.jpg (116.1 KB, 73 views)
File Type: jpg 15W-300b grid.jpg (120.4 KB, 50 views)
  Reply With Quote
Old 5th August 2015, 02:57 PM   #6
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Just for grins, ran it at 25W into 4 ohm load. Wanted to get a good idea of where the limiting factors are.

1.8% THD at the speaker terminals, significant clipping going on as indicated by the spray of high order and odds.

The signal at the 46 grids, however, is quite good, 0.01%. The input stage is happy getting pushed beyond what the amp is designed for.

The signal at the 300B grids is running into significant grid current, and the driver stage is running out of gas, 0.9%.

I may experiment a little to see at what point this hits the knee of the curve, but again, I have little intent to run the amp this hot. Perhaps modifying the bias conditions of the 300B stage could provide a little more headroom before the onset of grid current. It would be interesting to see how close to zero grid voltage I can run the differential driver stage before it falls apart.
Attached Images
File Type: jpg 25W.jpg (126.5 KB, 61 views)
File Type: jpg 25W-46 grid.jpg (118.9 KB, 46 views)
File Type: jpg 25W-300b grid.jpg (136.7 KB, 33 views)
  Reply With Quote
Old 5th August 2015, 03:32 PM   #7
diyAudio Member
 
SpreadSpectrum's Avatar
 
Join Date: Nov 2007
I'm assuming that these measurements are taken with a soundcard. What is the input bandwidth of the soundcard? What does the frequency response look like if you just test it in a loopback configuration?

I have found my cheap soundcard to be less than helpful for frequency response measurements. It just isn't capable of measuring outside the audio band. I use a signal generator and I sweep it.

The last amp I built has much wider bandwidth than the soundcard input both on the low and high end.

Edit: By the way, that looks like a very nice amp. Nice job.
__________________
My Blog: http://tubeswithatwist.blogspot.com/
  Reply With Quote
Old 5th August 2015, 03:40 PM   #8
diyAudio Member
 
SpreadSpectrum's Avatar
 
Join Date: Nov 2007
Also, do you have a setup to measure Zout? I just use a beefy op-amp to drive a 50 Ohm resistor which drives the amp output. Then I just measure voltages and calculate Zout (It is a simple voltage divider at that point).

That's another measurement to consider doing.
__________________
My Blog: http://tubeswithatwist.blogspot.com/
  Reply With Quote
Old 5th August 2015, 03:41 PM   #9
diyAudio Member
 
Join Date: Oct 2014
I don't know the frequencyresponse of your input transformer but the 46 is probebly limiting the response (low and high) too.

Quote:
Originally Posted by zigzagflux View Post
Frequency response and distortion. I know my distortion graph needs some help; haven't figured that out yet. Help is appreciated.

This amp will be driving speakers actively crossed over at the 80-120 Hz range, so I'm not too concerned about the low frequency behavior, but generally I'm pleased. -1 dB within 20 Hz - 30 kHz. Not too shabby for a no feedback design.
  Reply With Quote
Old 5th August 2015, 04:31 PM   #10
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Quote:
Originally Posted by SpreadSpectrum View Post
I'm assuming that these measurements are taken with a soundcard. What is the input bandwidth of the soundcard? What does the frequency response look like if you just test it in a loopback configuration?
I suppose additional info would be warranted. Using a M-Audio Profire 610, which is 24/192 with balanced in and out. Also have an active differential probe with 1/10 and 1/100 settings. Most measurements are taken 1/100 to protect probe and sound card. I have the Profire set to 96 kHz sample rate for all measurements. Loopback result with diff probe at 1/100 inside the loop is included. I lose almost 0.3 dB at 20 Hz and maybe 0.05 dB at 30 kHz, so not all that bad.

Also took loopbacks at the 46 grids, and another at the 300B grids. At the 46 I lose around 0.25 dB at 20 Hz, and maybe 0.5 dB at 30 kHz.

At the 300B grids I lose another 0.3 dB at 20Hz, and another 0.15 dB at 30 kHz.

Quote:
Originally Posted by SpreadSpectrum View Post
Also, do you have a setup to measure Zout? I just use a beefy op-amp to drive a 50 Ohm resistor which drives the amp output. Then I just measure voltages and calculate Zout (It is a simple voltage divider at that point).

That's another measurement to consider doing.
Not currently, but I could make one up. Got a schematic?

Quote:
Originally Posted by TrippleM View Post
I don't know the frequencyresponse of your input transformer but the 46 is probebly limiting the response (low and high) too.
Input transformer is not currently a part of the test system. It is only used when the source is unbalanced.
Attached Images
File Type: jpg Loopback.jpg (124.7 KB, 51 views)
File Type: jpg FR-46 grid.jpg (132.2 KB, 41 views)
File Type: jpg FR-300B grid.jpg (134.3 KB, 33 views)
  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
testing the 300b tube rubli Tubes / Valves 3 27th June 2014 12:46 AM
testing 300B on Hickok 539B/C tubelectron Tubes / Valves 19 28th November 2013 05:25 PM
300B Amp Testing mariner4 Tubes / Valves 11 22nd February 2012 03:13 PM
300B pp schematic found, what about it? jogas Tubes / Valves 24 28th August 2004 01:15 AM
Latest 6B4G PP and some thoughts about signal return paths... James D. Tubes / Valves 25 8th April 2004 10:46 PM


New To Site? Need Help?

All times are GMT. The time now is 06:00 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright 1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2