Go Back   Home > Forums > > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubelab Discussion and support of Tubelab products, prototypes and experiments

Reply
 
Thread Tools Search this Thread
Old 12th October 2016, 05:32 PM   #31
jdrouin is offline jdrouin  United States
diyAudio Member
 
Join Date: Feb 2012
Location: Tulsa, OK
This is fascinating stuff, Win. Wish I had more substantive things to add, so I'll just cheerlead for now.

Earlier you said the 307A is a find sounding tube, but have you listened to the amp in its current state (with the HV bench supplies and 6Xf rectifier) or just tested it?

A DHT SSE was going through my mind the other day as a curiosity.
  Reply With Quote
Old 13th October 2016, 04:04 PM   #32
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
307A:

I listened to it quite a bit through the test speaker - enough to know that it probably has the most potential of the three DH tubes I've tried thus far. I haven't actually tried it on the modified SSE board, however. The front end may or may not have enough gain.

A DH SSE has merit, I think - for someone. I'm just not sure if I'm that someone. I just really don't need another amp. I already have the TSE 801A, the child resistant SSE, a "6AQ5" P-P chinese amp, and an Elekit TU-879S at our lake house. On top of that, I have an unbuilt Elekit TU-870 amp and matching TU-875 preamp. So, I'm SE amp rich. And, I'm not allergic to silicon - I still have some genuine 1980's Nat Semi LM1875's needing a purpose.

I also have a set of P-P OPT's with a tertiary winding that I want to play with - maybe in hot rod SPP, many unbuilt HF transceiver kits going back to the 70's and up to the present DZ kit, and I STILL want to do another homebrew SSB/CW transceiver in a hybrid format, which strated this traninwreck diversion in the first place.

Quote:
A DHT SSE was going through my mind the other day as a curiosity.
Still, it's been said that 307A is a good 300B. Some have some real struggles with 300B in their TSE's. Assuming, without actually knowing as fact, that the front end has enough gain, A 307A SSE would not have the parts that seem to cause some TSE builders grief - the filament regulator, the adjustable negative bias supply, and the MOSFET's, but would introduce new issues of it's own. Cutting out the tube sockets and installing new ones is a pretty brutal mod. A tube socket adapter could be made, but it would be fiddly with the cathode wiring going in and out of it - and would add more height to an already tall tube(s). A separate filament transformer is needed for each power tube. You could go with DC on the cathodes, but then you start to get less simple. So, mixed bag.

I think enough of the merits to have hacked a board, and have thought about cake panning a proof of concept, but I'm not sure how much past that I would go with so many other projects needing attention. And, if I do that, I still might use 46 or 47 - haven't really decided. I have enough parts on hand to go either way, although the OPT's won't be that great.

To make a daily driver amp, to be usable at our primary residence, it would have to all be fit into a 12 x 10 inch chassis, and has to all fit under the existing Hammond cage. That is a difficult specification, given all the iron needed, but I've thought about it and have a plan in my head as to how it could be done, so who knows. Never say never.

Win W5JAG
  Reply With Quote
Old 27th October 2016, 03:48 PM   #33
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
For something a bit different- 3Q4, another directly heated pentode - something similar to this may make it into something else I'm working on.

This does a blistering 80 mW of output into an 8 ohm resistive load at first signs of clipping. This is the smallest tube OPT ( at least *I think* it's a tube OPT ) in my inventory. I was hoping it was 2.5K:3.2, but the 80 mW suggests it is 2K:8. 3Q4 needs 7.5K to 10K. The low end is - 1.5 dB at 300 Hz, which is good, the top end is about +1 dB at 100 Khz which is not so good and will have to be rolled off, a lot, but might make a good mosquito repellant as - is.

The input transformer is a transistor type, to make it easier to drive. I think it's 1K:10K. The square wave shows work is still needed. I think it was running about 50 volts on the plate when that was taken. The dh cathode is run off DC - the little LM317MP runs warmer than the tube ....

I think the next iteration of this will be push - pull, trying to use a small transistor type transformer to split the phase, and a small line transformer for the OPT.
Attached Images
File Type: jpg IMG_20161026_185043.jpg (485.7 KB, 227 views)
File Type: jpg IMG_20161026_185117.jpg (500.2 KB, 220 views)
File Type: jpg IMG_20161026_183624.jpg (391.0 KB, 215 views)
  Reply With Quote
Old 30th October 2016, 06:35 PM   #34
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
The mule outfitted as a push pull monobloc, the active, non power supply circuitry, being the SPP more or less, mostly more - a few component values changed to accomodate what I had in stock on a Friday night.

The power supply was moved into the back half, the hole for the filter cap was filed a bit and an octal socket was jammed ( literally ) into the opening. I removed the 60 ma choke and installed a much larger one of otherwise totally unknown specification.

The 12AT7 has one triode facing the input, the other facing the power tubes. That was about all the thought I gave to component layout.

I had a lot of trouble with the phase splitter. I wired the input triode - with a 10K plate load - measured the output with a VTVM, and it was fine. I wired the phase splitter triode, checked the outputs with a VTVM, and they were fine - near unity gain as expected. I connected everything up, checked for output, and got almost nothing. Checked everything again - lots of drive at the grides of the power tubes, virtually no output.

I disconnected one plate lead, and got a bunch of output. The OPT was cancelling the output instead of combining it. I got the scope and looked at the output of the phase splitter, and they were almost in perfect phase - off just enough to let a few milliwatts sneak out of the OPT.

After a couple of hours of struggling with something that consists of just two resistors, and getting absolutely nowhere, I was starting to feel like the poor bastard whose boomerang wouldn't come back. The problem was the 10K load resistor on the first triode, that I had picked to keep the gain down, as I have lots of drive available. I changed it to 100K, and bang, the following phase splitter worked as expected.

The power tubes are 6G6G, running about 7.5 watts each. This is more than double their rated dissipation, but they don't seem to care. This has also been my experience running them single ended. Cathode current is about 30 ma each side.

I just guessed at the feedback - I have a 2000 pf cap across a 5.6K resistor. It didn't seem to have much effect with the power tubes wired as triodes. It does seem to significantly lower distortion as power output increases in pentode mode. Frequency response seems unaffected either way, it's within about -2dB from 40 Hz or so to almost 50 kHz. Power output is about 1.25 watts before the first trace of sine wave distortion. B+ is about 260 volts loaded.

The OPT is the Atlas Sound HT87 line transformer. P-P 6G6G's want to see 15-20K plate to plate, the Atlas is 5K, so, way off. Performance would no doubt improve with the correct load.

Wn W5JAG
Attached Images
File Type: jpg IMG_20161030_000534.jpg (412.4 KB, 212 views)
File Type: jpg IMG_20161030_000126.jpg (423.7 KB, 208 views)
File Type: jpg IMG_20161030_012915.jpg (290.4 KB, 80 views)
File Type: jpg IMG_20161030_013238.jpg (242.9 KB, 41 views)
  Reply With Quote
Old 8th November 2016, 03:16 AM   #35
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
SV811-10

I modeled the output stage in SE Amp Cad, but preliminary resuts were disappointing on the SSE, for several possible reasons. This could be hard. More work / experiments are needed. It was killa on the TSE board, at least before it set my board on fire. Maybe the zero bias transmit tubes could be persuaded to work here.

45

Worked pretty close to as modeled by SE Amp Cad. Produced more than a watt with a 5K transcendar as the load. Practically plug and play once you move past the drama of hacking out the existing power tube sockets. This would probably be a good amp for someone who wants a really simple DHT. The short bottle 45, and the stubby 5930, would fit under the Hammond cage on the CR SSE, making it easier to do a child resistant DHT.

2A3

I didn't have any at the house, so I haven't tried it. I'll try to bring home some 5930 next time I'm at the warehouse of junk, and try them. I expect them to be plug and play, like 45.

The red spots on the paper are an artifact of how tight a wavy washer holds these sockets to the board. There was a lot more than that, but it missed the paper.

Win W5JAG
Attached Images
File Type: jpg IMG_20161104_210507.jpg (369.5 KB, 69 views)
File Type: jpg IMG_20161104_215952.jpg (330.0 KB, 64 views)
  Reply With Quote
Old 11th November 2016, 03:01 PM   #36
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
I changed the line transformer to a Stancor 10K plate to plate, which is still probably a little low, but a better match than the HT87 was.

I didn't take any notes, but IIRC, it was -2 dB at about 40 or 50Hz, and maybe - 1 dB at 20 KHz, so pretty well full range. Output is about 3.5 - 4 watts. Each tube in the push pull pair is running about 7 watts, so that seems like a reasonable enough output number, but I'm not really familiar with push pull amps.

I stuck a barrier strip on the "chassis" where the small choke used to be, and using some scrap pieces of wire, I brought the output of the first triode amp and grids of the power tubes to the barrier strip, also a ground connection. To make it easy to attach clip leads, I used scrap pieces of component ends and bent a little circle into one end, for the clip lead to get a good grip on. The straight end goes to the barrier strip. Waste not, want not.

This way I can easily use the output of the triode first stage, or the phase splitter, or the push pull power stage, as known good building blocks for working on other projects. The input tube could be pulled, and a differential input can easily be applied to the grids of the power tubes, to use just the power stage.

This is purely for convenience - the chassis could be flipped over to do all of this with clip leads - this just makes it a lot easier, neater, and faster. A blocking capacitor is needed for the output of the triode first stage, otherwise there would be a lot of DC at that terminal of the power strip. The power tube grids are already isolated from the splitter by the coupling capacitors, so bringing the signal out from the grids takes advantage of that. Feeding a signal into the power stage from the external strip would require blocking / coupling capacitors however.

The obvious first use here of the phase splitter outputs / inputs is making an SSE board push pull. I do recall at least one thread a few years ago on this. Should be easy, really - making one into a hot rod or top fuel psuedo SPP.

Also, it seems that this converts the mule into a tube tester of sorts. Since the output of the first stage is easily accessible, a signal of known ac voltage can be fed into the input, then measured at the output, and the gain of that section can be measured, also the noise, and since the AA-1 is quite sensitive, the IM distortion characteristic. Likewise with the phase splitter half, although I'm less sure about the validity of the data to be had there. More experiments will be needed, but it looks like it should be an easy way to match tubes for similar characteristics.

Since I had one pin on the barrier strip left over, I brought the cathode of one of the power tubes out to it. Measuring the cathode voltage here allows one to calculate the current, thus power tubes can be matched for current this way. While the onboard power supply is weak sauce, the external power supply can be easily plugged in to set whatever voltage is needed for the tube under "test". 6G6G uses the same pinout as the common stuff like 6V6, 6L6, etc., so it "should" be quite useful for this.

Anyway, Jeff, if you're still reading, thanks for the great idea, and sorry I've threadjacked your thread and trainwrecked it so badly!

Win W5JAG
Attached Images
File Type: jpg IMG_20161110_234112.jpg (422.5 KB, 74 views)
  Reply With Quote
Old 11th November 2016, 03:27 PM   #37
jdrouin is offline jdrouin  United States
diyAudio Member
 
Join Date: Feb 2012
Location: Tulsa, OK
Hey, I'm still reading! Thanks, and no worries about the threadjacking. This is really fascinating and I've been watching it closely. I just don't have enough knowledge to chime in with.

One thing I've come to realize is that I prefer the sweet, warm, open, vintage sound that tends to get eliminated by ICs. I might try removing the MOSFET source followers but retaining the CCS chips, as originally suggested in this thread, just to see if it changes the sound. Also adding better coupling caps than the Auricaps that are in there now (like Mundorf Mcap silver-oils, maybe).

But in the spirit of frugality that you've inspired, I'm more interested in breadboarding a new 300B amp with the equipment I already have. Maybe I'll use a diode bridge in the PS section, but other than that it will be a non-transistor 5842-->300B with a double pi filter in the PS and the 5K:8ohm:15W Electra-Print OPTs, plus my existing PT. Maybe with a traditional circuit like the Williamson. I want to let the second-order harmonics sing!

Then I'll reassess. I once heard a 6SN7-->300B about 15 years ago that was revelatory. So, if the above project works, and I'm still wanting something different, then I'll invest in some new iron and a pair of 6SN7s and make a pair of mono blocks. I keep coming back to the 300B design in the Fi Primer and just want to try it.

Now that I've made and modified the TSE, I have enough knowledge to attempt it, and for that I thank George.
  Reply With Quote
Old 11th November 2016, 04:13 PM   #38
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
That preamp build you just finished looks terrific.

My metal working skills would have to improve a lot just to get to the point it could be said they really suck.

I take it you had no ground issue(s) to resolve with your split chassis build? I have a lot of old transciever supplies I've thought of pressing into audio amp service, but a dedicated external supply just for audio amps would have a lot of virtues. I've got the parts to build a pretty stiff one, just haven't done it.
  Reply With Quote
Old 11th November 2016, 05:06 PM   #39
jdrouin is offline jdrouin  United States
diyAudio Member
 
Join Date: Feb 2012
Location: Tulsa, OK
Thanks! All I used was a step bit and various sizes of regular drill bits to make the holes. Oh, and a Black & Decker rotary tool with the Dremel #426 cutoff disk to make the rectangular opening for the IEC inlet.

The rest is Rustoleum all-surface spray paint (matte French cream).

The grounding is very strong. All the power conductors are 20 awg but I used 18 awg for the chassis grounds, and kept the wider bandwidth in the umbilical by soldering them to both the ground conductor and the cable shielding. There is a barely noticable 60hz hum but that's because the outlet it's plugged into is not grounded. I'm going to try the old "ground the outlet to the box" trick to see if that works.

300B ideas aside, I think my next project will actually be to make a line preamp, either with the 26 or 10Y tube, so I can drive mono blocks. The 26 is especially susceptible to hum, so I'd probably do a two-chassis build for that too.

It's more complicated and time consuming to pull off, but it's worked well so far.
  Reply With Quote
Old 11th November 2016, 05:34 PM   #40
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Quote:

a non-transistor 5842-->300B with a double pi filter in the PS and the 5K:8ohm:15W Electra-Print OPTs, plus my existing PT. Maybe with a traditional circuit like the Williamson. I want to let the second-order harmonics sing! ...
I wonder if there is enough gain out of the 5842 to drive the 300B to clipping.

An interstage transformer to boost the drive voltage might be useful, but I don't know much about them, vis-a-vis hi fi.

I have one, with adjustable ratios, but just haven't got around to experimenting with it. That SV811 on the SSE board is a likely first place to try it.

My four year old is at my office today, so .... work ..... is futile .....

Win W5JAG
  Reply With Quote

Reply


Tubelab SE: Removing MOSFETs?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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Mosfets for Tubelab SE? mctavish Tubelab 60 14th April 2018 04:50 AM
Tubelab SE zxx222 Tubelab 172 4th September 2016 12:53 PM
F/S Tubelab SE stonehome Swap Meet 14 25th December 2014 10:56 PM
Tubelab SE vs Simple SE Pez Tubelab 25 29th October 2011 02:16 AM
Need Help with Tubelab SE lth1 Tubelab 9 1st October 2010 09:06 PM


New To Site? Need Help?

All times are GMT. The time now is 12:59 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki