801A in Tubelab Single Ended? - diyAudio
Go Back   Home > Forums > Commercial Sector > Vendor Forums > Tubelab
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs 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 2nd June 2011, 03:47 AM   #1
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Default 801A in Tubelab Single Ended?

Anyone tried these in the Tubelab single ended?

A big lot of ex FAA tubes was given to me a few months ago. In going through the lot I see some 801A's. Aside from the 7.5 volt filament, these look like they should drop right in to the TSE, and should easily outperform 2A3 ( which isn't saying much, 5930's are all I really care for in this family ) and be as good or better than 300B, and 307A / VT-225.

The thoriated tungsten filament should look good in the amp. My Tubelab SE has been down for rebuild for quite a few months now, really ever since I veered off on the 6146 tangent, and I'm thinking when I put it back together I might try these tubes in it.

Win W5JAG
  Reply With Quote
Old 16th November 2015, 01:43 AM   #2
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Who doesn't love thoriated tungsten?

Win W5JAG
Attached Images
File Type: jpeg 801_lights_on.jpeg (12.7 KB, 367 views)
File Type: jpeg 801-light_only_.jpeg (12.7 KB, 367 views)
  Reply With Quote
Old 16th November 2015, 04:05 AM   #3
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Well, those pictures sucked. Maybe these will be larger.

The first picture is with the room lights on. I made a few changes ( and blew up a part after that pic was taken ). The second is with the lights out; all the light seen is coming from that single 801A - you can see the 801 actually casting shadows.

I can see, I think, how this is going to go: Step 1 was getting past the 7.5 volt filament hurdle.

Step 2: shifting the range of bias voltage to make it more suitable for the 801.

Step 3: These tubes are going to need B+ voltage beyond what a normal TubelabSE would see. The upper end of the typical 300B B+ voltages is probably going to be about the minimum that 801 wants to work at.

When I first powered it up, it was running around 200 volts on the plate - I was using the same small PT that I used to bring the board back to life after it's long hibernation, and make sure everything was working as it should. I had a 45 on the board for that.

With a 7K load, the 801 didn't sound good at 200 volts. At 300 volts, about as much as I can wring from that little PT, and after listening for awhile, I found myself really liking it. I think I liked it better than the 45, although, in fairness, I kept the 45 down around 200 volts. From looking at the data sheets, I'm not pulling as much current as I expected at a given bias point, so I'm not yet sure what is up with that.

Win W5JAG
Attached Images
File Type: jpg WP_20151115_20_16_02_Pro.jpg (929.6 KB, 364 views)
File Type: jpg WP_20151115_20_16_28_Pro.jpg (464.9 KB, 348 views)
  Reply With Quote
Old 1st December 2015, 03:30 PM   #4
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
First, the disclaimer:

The TSE was not intended for operation with 801 tubes. Anytime a product is made to do something other than the intention of its designer, bad results are possible. A one off modification may, or may not, be repeatable. Some choices and experimentation will be required. Some choices and experiments may lead to bad results including, but not limited to, component damage. Anyone embarking on modification of their TSE board must be aware of, and accept the possibility of, such bad results.

In other words, just don't do it.
  Reply With Quote
Old 1st December 2015, 03:40 PM   #5
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
The TSE board was designed for operation with 45/2A3/300B tubes using directly heated oxide coated cathodes at up to 6.3 volts.

The 10/50/801 family of tubes uses directly heated thoriated tungsten cathodes at 7.5 volts. The thoriated tungsten operates at a much higher temperature and glows with a brilliant white light, much like a small electric light bulb.

The regulator on the TSE board is specified for a maximum voltage output of 5 volts. It MAY be possible to use this regulator as part of the conversion process by elevating the chip a few volts above ground so that it only sees 5 volts across it, when the output voltage is actually higher. I didn't try this. I decided the simplest solution was to remove the regulator from the board, together with one lead of R2. This allows easy access to the cathodes of the power tubes.

AC versus DC

The AC versus DC decision is a choice that must be made to proceed with the conversion. Some audiophiles say AC sounds better on the cathodes of DHT tubes, and that this is especially true for thoriated tungsten tubes. There is a resistance in the cathode, so with DC applied, there will be a temperature difference from one end of the cathode to the other that would not exist if AC was used. There is a big potential for, maybe a certainty of, low level 60 and 120 Hz IMD products in the amplified output when using AC.


I went with DC for several reasons: First, the designer is a practicing engineer and I'm not. Second, the RCA 801 data sheet curves are all specified at DC on the cathode. Third, the use of DC allows the existing filament/cathode supply to be used as designed, largely intact. Fourth, there are no provisions on the TSE board for hum reduction components if AC were to be selected; it was not readily apparent to me that there is a convenient way of installing them.


In short, DC seems likely to produce the best result, and presents the fewest practical problems in the process.
  Reply With Quote
Old 1st December 2015, 03:56 PM   #6
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
The Voltage Situation

The data sheet specifies 7.5 volts so that would seem to be that. It does not specify a tolerance. In doing some casual internet reading on this, most of the ( anecdotal ) information on this subject is from experience with, or in some instances actual manufacturer data, on the thoriated tungsten filaments of bigger RF power tubes. It goes something like this: the tolerance is typically specified on these tubes as + - 5 per cent. Operation above this tolerance is really bad and will lead to shortened tube life, and operation below this tolerance is really bad and will lead to shortened tube life, except that there is an undervoltage condition, of 3 or 4 per cent less than nominal, where tube life is greatly extended. The anecdotal evidence of this with big transmitting tubes comes from both broadcast and amateur operations, and is common enough that I accept it as true. I have no idea if any of this applies to small RF tubes like 801, or, if it even does, whether AC or DC was being used on the cathode. I suspect AC.

There is also the "starved triode" school of thought that suggests that with small and large ish DHT's, performance improves dramatically at reduced cathode voltages, as evidenced by actual measured data reported here by Steve Bench:

DHT with starved filaments.


Further, Steve Bench postulates that there is a relationship between power drawn by the cathode, and permissible current drawn by the plate, and that cathode undervoltage need not, and will not, lead to shortened tube life, so long as plate current is reduced in some proportion to the reduction in cathode power:

Heater / Filament Efficiencies


His site looks to be long extinct, but is mirrored on several other sites, and there looks to be no follow up to support or refute this proposition.
  Reply With Quote
Old 1st December 2015, 04:17 PM   #7
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
To proceed, I removed the 2.5 volt jumpers previously installed on the board, and put the 6.3 volt transformer jumpers in place on the board. I soldered a jumper across the pads on the back of the board that would normally be used for input and output of the regulator chip.

In theory, a 6.3 volt AC output should rectify out to well over 8 volts. In practice, once the outout is loaded down, it's a lot less. My power transformer is significantly oversized for a project like this, so it is very lightly loaded by the current requirements of the 801A's and 5842's.

With both 801A's in place, I measure about 7.3 volts DC on the cathodes at initial turn on, which drifts down to about 7.1 volts over time. It appears that as the 801A's warm up, the cathodes draw more current until they stabilize at operating temperature and current. This is a smidgen below the lower tolerance limit for 801A, but probably within the margin limits of my measurement devices. I measured 6.0 volts on the 5842 filaments.

I doubt that in practice, this will cause any real difficulty. The color looks normal.
Attached Images
File Type: jpg IMG_20151130_234951.jpg (644.7 KB, 296 views)
  Reply With Quote
Old 1st December 2015, 04:34 PM   #8
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
The Bias Supply

When I was first testing the 801's at lower plate voltages than they are intended to operate at, I was concerned that the TSE bias supply would need some rework, as I couldn't move the bias voltage positive enough to get the current draw I was seeking. I think there are a couple of reasons for this: one, the plate voltage was too low. Two, the cathode voltage was low enough ( 6.7 volts ) that, combined with the low plate voltage, it just needed less bias than the board could nominally supply.

My TSE board was originally built for 45's. My power transformer has a separate screen voltage winding that produces dc output voltages in the mid - low 250's when used with a circuit similar to that of the bias supply on the TSE. To avoid the inconvenience of changing out the dropping resistors on the TSE board, it seemed logical to use the screen voltage taps. The bias rectifiers are nominally fed by the HV secondaries on a stock TSE board. To access these rectifiers independently, the easiest and most convenient way is to simply lift one leg of each diode from the board, and solder the lower voltage secondary directly to the leads of the diodes.

I measured the range of bias voltage at each socket as being approximately -30 to -80 volts. Once the 801A's were being fed with proper plate voltage ( I have begun at 380 volts ) this range is adequate to properly bias 801A.
Attached Images
File Type: jpg IMG_20151201_092558.jpg (320.1 KB, 78 views)
  Reply With Quote
Old 1st December 2015, 04:52 PM   #9
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
The 801A needs high ish voltage for a small ish tube, and it also needs a high plate load impedance, 10K or greater.

The only chassis I had readily available that could meet these requirements was the venerable SSE board that has seen countless hours of use and misuse. As an additional bonus, it had the lower voltage secondary available for the bias supply.

Indeed, I think this chassis has the capability for SV 811 or SV 572, although these bottles may be too big to fit the spacing on the TSE board.

Anyway, the SSE board, screen supply, and unnecessary bits were removed from the old chassis, and on went the TSE board. A 47 uf 450 volt cap was installed as the first cap in the B+ supply, and some other bits I had blown up were replaced before the board was installed.

If it looks like it got wired in a linear fashion, that's because it did. Although I had already re verified he board to be in proper working order ( less the blown bits ), I proceeded one step at a time in an abundance of caution, testing as I went. It took about three hours from beginning dismantlement to listening to the 801 TSE on the new chassis.

First, I wired in the Hammond 125ESE OPT's and set the load for 10K.

Next I wired up the 7.5 volt and 5 volt secondaries, and individually tested each for proper operation before proceeding.

Next the bias secondary was wired up, and range of voltage at each socket was measured. Bias was set to maximum negative prior to installation of the power tubes.

Finally, the high voltage secondary was wired in, and B+ was confirmed to exist at the power tube sockets. For initial power up and testing, I chose a 5W4GT rectifier. It can easily handle the current requirements of 2 5842's and 2 801A's.
Attached Images
File Type: jpg IMG_20151130_224343.jpg (507.5 KB, 75 views)
File Type: jpg IMG_20151130_224432.jpg (422.6 KB, 81 views)
  Reply With Quote
Old 1st December 2015, 05:14 PM   #10
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Initial power up was uneventful, thankfully. The bias had already been set to maximum negative on each power tube.

I installed the 5842's and verified they were drawing current on the panel meter. I connected a speaker to the right channel, installed the right power tube, and adjusted the bias for an additional 40 ma on the meter. Thoriated tungsten tubes are "instant on" devices. I jumpered an audio source to the right inputs, and verified that music in produced music out at the test speaker.

I placed a 10 ohm dummy load across the left speaker and installed the left power tube. I adjusted the bias on the left channel for another 40 mils on the panel meter.

I then removed power, cooled everything down, and swapped the speaker and dummy load to verify that music in produced music out at the test speaker on the left channel.

Then, I took a few more measurements and just listened. Until 1:00 AM. With the 5W4, B+ at the power tube pins is 380. Bias is, IIRC, -40 ish. Cathode voltage is noted above. Eventually, I will work it up to about 440 volts, the limit of my filter caps.

The inputs remain to be wired to the board. I am undecided on metering. These are trivialities.

I have read the 801A sounds thin. That is definitely not true as installed in the TSE. Even on my crappy test speaker, the 801 TSE has the robust dynamics of a 6L6 and the detail of a 45. I only expect it to get better as more voltage and current is applied.

Win W5JAG
Attached Images
File Type: jpg IMG_20151130_233330.jpg (333.2 KB, 101 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
Single tube, Single ended, 20W, impossible? dannydanger Tubes / Valves 35 3rd April 2012 01:19 AM
Tubelab Single Ended xecluded Tubes / Valves 14 6th April 2011 11:32 PM
Tubelab Single Ended Amps Structo Tubes / Valves 9 22nd December 2010 04:13 PM
Single Ended, Single driver: In-room distortion measurements Gerrit Boers Full Range 4 27th November 2007 08:12 AM
Anyone play with the Zen Single-Stage Single-Ended Class A? Rino odorico Pass Labs 13 29th August 2002 09:01 AM


New To Site? Need Help?

All times are GMT. The time now is 02:10 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki