Go Back   Home > Forums > > >

Tubelab Discussion and support of Tubelab products, prototypes and experiments

Reply
 
Thread Tools
Old 14th September 2021, 10:03 PM   #61
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Those do not look like the right pots either. The Bourns 3309P pots that I have been using for 10 years went extinct recently. I have been using some Piher 10 mm pots that fit fine. I'm digging through the pile of part on my bench for the empty bag to get the number. I'll post it as soon as I can find it.

I found the empty bag. The MFG's part number is PTC10LV10-104A2020 but Mouser and Digikey are out of stock. I'm looking for alternatives.
__________________
Tubelab, I blow stuff up so that you don't have to.

Last edited by Tubelab_com; 14th September 2021 at 10:09 PM.
  Reply With Quote
Old 14th September 2021, 10:47 PM   #62
Francois G is offline Francois G  United States
diyAudio Member
 
Join Date: Jun 2004
Location: Illinois
UNSET Beta Board Build
Thanks, George! I see now that the trimmer I mentioned in post #60 seems to be a “finger adjust”.

The item you mentioned above is currently out of stock at Mouser, but they expect plenty to arrive by 9/24/21.

PTC10LV10-104A2020 Amphenol Piher | Mouser

Last edited by Francois G; 14th September 2021 at 10:49 PM.
  Reply With Quote
Old 14th September 2021, 11:10 PM   #63
spiggs is offline spiggs  United States
diyAudio Member
 
Join Date: Dec 2016
Location: Huntington Beach, CA
UNSET Beta Board Build
Digikey has stock of the ones I used Blocked
  Reply With Quote
Old 15th September 2021, 12:36 AM   #64
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
The Digikey part mentioned by Spiggs is rated for 150 mW (the PT series). With the resistor values stated in the parts list it runs at about 100 mW, so I chose the PTC series which is rated for 330 mW. Mouser has some PT series parts in stock too, the PT10LV10-104A2020-PM-S.

Users including myself have found that setting the bias to an exact current with a single turn pot is hard to do. I chose the resistor values for R9 and R111/R211 to provide a pretty wide adjustment range so that a wide variety of tubes and supply voltages can be used with this board. These values can be increased to narrow down this range once tube selection is made, making adjustment easier and reducing power dissipation in the pot.
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 15th September 2021, 05:34 AM   #65
spiggs is offline spiggs  United States
diyAudio Member
 
Join Date: Dec 2016
Location: Huntington Beach, CA
UNSET Beta Board Build
Did some measurements using a better multi meter with fresh batteries plus a bias increase. Here is where it landed tonight.

B+ 477V
R115/R215 450V
cathode 70V
bias at 65mA

So that puts 380V across the tube and 24.7watts dissipation

Probably won't have time for the next few days to swap the speakers between the 4 and 8 ohm taps and take a good listen but that should be fun.

Interestingly I was looking at the thread "Those Magnificent Television Tubes" and saw mention of the "winglet" Sylvania 6DQ5 style tube being good for a bit more than most. I have a pair of these in the UNSET currently. Perhaps I could bump the bias up a bit more and see how it goes.
  Reply With Quote
Old 15th September 2021, 10:46 AM   #66
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
I have heard Smoking-Amp talk of such tubes on these forums for a while. The pictures of your board was the first time that I have seen one. That reminded me of a big box of used sweep tubes that I have been collecting, so I did some digging and found one. I'll test it the next time I have the opportunity to rewire my board for the 6DQ5 type tubes.
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 16th September 2021, 04:42 PM   #67
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
One of the most common questions I get about my amps, other peoples amps, and tube amps in general is "how many watts" or "how much power can I get?"

The answer is always a guess that requires more information than is usually provided. There is usually one sweet spot for a given tube (limited by plate dissipation in an SE amp), the supply voltage, and the OPT.

I set up my UnSET board with a Toroidy 1500 ohm OPT which can be run at 1500 or 3000 ohms. I made a chart of real measured data for both impedances. I used a 26HU5 because I have a few, and it can be used up to 50 watts for short periods of time.

For a given tube choose the maximum plate dissipation you want to run it at, and only choose combinations below that number in the "Plate diss" column. From there pick a supply voltage you can make to arrive at the best OPT load impedance, or pick an OPT and look for a convenient OPT.

You will see that the maximum efficiency coincides with symmetrical clipping. You want to be close to this point if possible.

From these charts I can see that I can get about 25 watts from the 1500 ohm OPT on 450 volts if I beat the tube hard at just over 41 watts.

Getting 25 watts with a 3000 ohm OPT needs between 42 and 46 watts on 550 volts.

It looks like my best choice is 1500 ohms, probably somewhere between 400 and 450 volts and somewhere between 100 and 110 mA.

Maybe I can push a bit more dissipation into the mosfet. Note that raising the screen voltage shifts more dissipation to the fet since more negative bias is needed.
Attached Files
File Type: pdf Power Output 1500 ohm OPT.pdf (368.7 KB, 26 views)
File Type: pdf Power Output 3000 ohm OPT.pdf (370.6 KB, 25 views)
File Type: pdf Power Output 5000 ohm OPT.pdf (365.6 KB, 19 views)
__________________
Tubelab, I blow stuff up so that you don't have to.

Last edited by Tubelab_com; 17th September 2021 at 12:53 PM.
  Reply With Quote
Old 17th September 2021, 08:50 AM   #68
sunil is offline sunil  India
diyAudio Member
 
Join Date: May 2003
Location: Bangalore, INDIA
UNSET Beta Board Build
Wow, thats close to 50% efficiency

In the charts, a total Diss of 24w is ~ 300 and 350v B+ (symmetrical clipping). Would this apply for the smaller tubes like EL34. Tubes in group 2 & 3 are pretty hard to get here but I have some vintage Mullards and Telefunken El34s.

Would those be ok? Shows around 45% efficiency and 11 - 13w.
  Reply With Quote
Old 17th September 2021, 01:23 PM   #69
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
I added the Edcor 5000 ohm OPT data to post #67.

Yes, some of the efficiency numbers are beyond the theoretical 50% maximum for a conventionally configured single ended output stage. There are two reasons for this. Measurement error, and the fact that this is NOT a conventionally configured SE stage.

I set an arbitrary 5% THD point for all measurements. This is usually the point where clipping becomes obvious on the scope, and where things start to sound "turn it down" kind of bad. All my dissipation measurements rely on low budget meters, and a 50 year old Fluke 407D power supply that is completely original. It does drift a little over time and with line voltage fluctuations. Overall I would say that the total measurement error is considerably less than 5% of reading, so the calculated efficiency is correct to within 1 or 2%.

The CED circuit drives the cathode while holding the screen grid constant and grounded for the audio signal, and also keeping the control grid relatively constant while also applying direct NFB from the plate.

Audio power is applied to the cathode, as in a grounded grid amp. This power flows directly to the plate, boosting the apparent efficiency.

This used to be a neat trick in the old days of ham radio when hams were restricted to a certain DC power input to the tube, rather than the maximum output power regulations of today.

The same thing happens in the CED circuit. The plate to control grid feedback makes the pentode work like a triode with it's grid grounded, and the power added to the cathode appears at the plate with the power made in the tube itself. This can result in a real over 50% efficiency number.

As Zintolo found in his simulations, changing the ratio between tube and fet can boost this to extremes in a computer sim, but real world limitations like mosfet SOA, limit things. For now I have been running 10 to 12% of my total dissipation in the fet. Further experiments are being done to find more options.

Back in the early days of the CED I did experiment with the KT88. It does not have the sensitive screen grid that the TV sweep tubes do. The EL34, 6L6GC and most other typical audio tubes are similar. They do not respond as well to screen drive that TV sweep tubes do, so I'm not so sure how they will work out in the UnSET, as I have not done much work with them. They will obviously need a screen supply with more than 190 volts, and probably a different divider ratio on the control grid. I will get to these experiments eventually.
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 17th September 2021, 03:52 PM   #70
sunil is offline sunil  India
diyAudio Member
 
Join Date: May 2003
Location: Bangalore, INDIA
UNSET Beta Board Build
Thanks George. I may get 6DQ5 and think it is going to be fine.

I'm building two SSEs with 807 | 6V6 (280 - 320v) on one and EL34 | 6P3S (400+) on the second. That should take care of EL34 sound

The UnSET is after that
  Reply With Quote

Reply


UNSET Beta Board BuildHide this!Advertise here!
Thread Tools

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
UNSET is coming? Tubelab_com Tubelab 302 8th September 2021 09:14 PM
If UNSET and the RCA50W Had a Baby SpreadSpectrum Tubes / Valves 35 23rd June 2021 12:02 PM
CED/UNSET/CORONA-like ECC88 hybrid cascoded driver zintolo Tubes / Valves 20 24th March 2021 01:18 PM
First sim of UNSET circuit jderimig Tubes / Valves 12 20th February 2021 12:17 PM
UNSET questions jderimig Tubelab 0 18th February 2021 07:37 PM


New To Site? Need Help?

All times are GMT. The time now is 03:08 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 13.64%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki