The AM Forum
March 28, 2024, 05:53:08 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: 813 plate Z  (Read 12619 times)
0 Members and 1 Guest are viewing this topic.
kc2ifr
Guest
« on: January 04, 2007, 07:02:17 AM »

What does triode connecting an 813 do to its normal plate impedance?
Thanks,
Bill
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #1 on: January 04, 2007, 11:53:41 AM »

Well !
Depends on how you connect the grids and the plate for two different types of triode - You can get high mu or low mu. (gain figures)   Typically hi mu is 100 or so.  Lo mu may be 10 to 20.  Plate resistance vaires correspondingly.

I'll leave that as an excercise for the student.  Many sources on the web.

Hint. For a typical tetrode converted to a triode; not necessarily good for beam focused tubes like the 813, 6AQ5, 6L6, etc.

Connecting the screen grid to the plate and leaving the control grid separate gives one sort of gain.
Connecting all the grids together as one and leaving the plate separate gives another gain. 

Remember that mu = gm * Rp, so there's your plate impedance  ---  Rp = mu/Gm
Transconductance, gain and plate impedances for a wide variety of operating conditions are given in lots of tube tables including the ARRL manuals of yore.  Some manuals gave gains for both tetrode and triode connections
Yes, back in the day when processed vacuum ruled.
Logged

RICK  *W3RSW*
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #2 on: January 04, 2007, 12:34:04 PM »

I hook all 3 grids together for triode connected 813's...In real world use I have found that they are not all that sensitive to impedance matches when used as modulators in the voltage range of 1800 to 3000...I have several rigs in use with triode connected 813 modulators using whatever ratio mod iron I could find......modulating triodes, pentodes, and tetrodes and they all perform very well.I have seen published figures of 12000 ohms to 21000 ohms as the plate to plate impedance of 813's so there seems to be some difference of opinion there as well.   good luck,  Steve
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #3 on: January 04, 2007, 01:27:45 PM »

Hi Bill,

As pointed out above, the question does not necessarily have a specific answer.  Gain devices such as tubes or MOSFETs, etc. don't necessarily have a specific impedance (resistance) into which they must work.  The ability of any device to work into a specific resistance is dependent on the current and voltage capabilities of the device, and the gain needed (or available).  The specific voltage gain is dependent on the actual device's gain characteristics (transconductance, in this case), and the load to which it is connected.

So, now to the the question: - will an 813, triode connected, be capable of working into a lower impedance (i.e. more current) than the tube same connected as a pentode.  All this assumes linear (or audio), distortion free operation.  Most likely yes, if the 813 is like other pentodes/tetrodes.  The voltage gain is significantly reduced - in fact, depending on the tube, reduced to very low numbers (not sure about the 813 specifically).  The grids will be driven positively with respect to the cathode, requiring a low impedance driver - this driver may also be required to deliver considerable voltage.

The linearity of the tube as a voltage/power amplifier may or may not be satisfactory.  I have had varying results using multi-grid tubes as triodes.  The driver design can be critical, as some tubes' grids exhibit a widely varying load, depending on the specific input voltage.

You probably have some 813s lying around, otherwise the 810 would be a better choice if you want a triode, assuming a similar power output.

So, what are proposing to build ?   Cool  We're all ears (Ross Perot).   Grin

Regards (and happy year year!),

Steve
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
WB3JOK
Member

Offline Offline

Posts: 637



« Reply #4 on: January 04, 2007, 02:01:47 PM »

So, what are proposing to build ?   Cool

Oh goody, another homebrew amp project... let's make some smoke  Cool

Quote
We're all ears (Ross Perot).   Grin

Not quite the season yet but that reminds me of this Easter bunny cartoon:


 Grin
-Charles
Logged
kc2ifr
Guest
« Reply #5 on: January 04, 2007, 05:18:19 PM »

You guys are great!!!! Cheesy
I have a modified Johnson 500 that uses 572b's as modmulators. They have gone soft and I cannot find any good-ole american made replacements and I WILL NOT USE the CHINESE version. SOOOOOOO I am going to replace them with 813's triode connected. They seem to be a good choice. Zero bias, no screen voltage and will like to 2000 volts the 500 will supply. I will drive them either with a backward connected Hammond outpoot tranny or a cathode follower. I now have a cathode follower driver that was built by Guy...W1FRM using 6L6's but not sure it will have the monkey swing for the 813's.
Anyway, that is the plan. I was worried that the 813's would not like the 500 mod iron but I think it will be ok.......what do u think?Huh

Bill
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #6 on: January 04, 2007, 08:05:07 PM »

Hi Bill,

I've never actually worked with a 572B, however in looking at the curves, the 572B *appears* to be a slightly higher power tube than the 813.  However, the difference may not be an issue.

The 572B appears to have a very high peak current capability.  With +100 volts on the grid, you will be pulling close to an ampere of current.  That's pretty good.

Looking at the curves for the 813, to achieve a similar current swing will involve driving the screen to at least +300 volts, and the control grid to +100.  At +100, the grid will pull about 40 mA of current.  The grid #1 dissipation is not specified in my documentation.  You can probably figure an average grid #1 current of 20mA, and you probably won't burn it out.

I seem to recall that the 813 is reasonably well screened, so grid #2 will have a great affect on the plate current.  I would figure on driving grid #2 to at least +300 volts.  The grid #2 dissipation is 20 watts.  I figure 150 or 200 mA of peak grid #2 current - maybe a bit more.  As long as the average dissipation is 20 watts or less.

It would be impractical to drive the grids from different sources, but you COULD drive them in such a way as to make the grid #2 voltage higher than the grid #1 voltage, and keep this voltage constant.  You might consider some experiments in this area.  Drive the screens directly, and use a zener diode (or a couple of them) to drop the grid #1 voltage (bypass this with a capacitor).  You might try a 100 volt zener, or a 150 volt or a 200 volt.

So, the zener would connect between the screen and the control grid, and will cause a constant voltage difference to exist between the grids.  Both grids will then be driven, but will be separated in voltage.  You will need a pull down resistor at grid #1 to force the zener to reach its knee, and maintain the voltage difference.

Looks as if a 600 volt P-P audio swing might be in order, to leave sufficient headroom.  Actual operation may be less.  You should have plenty of driver current available.

All of this is THEORITICAL, as I have not actually implemented this... however, looking at the curves, it makes sense.

It may or may not work out if you simply tie the screen and control grids together.  I would try the zener arrangement (bypassed with a capacitor).  The screen grid needs a lot more voltage swing, and also DC potential, to have control over the electron stream, as this grid is far from the cathode.  It will also draw less current per given voltage.  The control grid, being very close to the cathode will consume a lot of current, and the electron stream may not be sufficiently excited due to the relatively low screen voltage, so achieving a large plate current swing may be difficult.  Distortion will most likely result.  Because the grid current/voltage characteristic is non-linear, a resistor is not a good choice for dropping the grid #1 voltage.  A zener will force a constant voltage difference between screen and control grid.

Sure, it may "work" if you just tie the grids together and drive them, but you may not be satisfied with the results, distortion-wise.  Albiet, I am very picky about this sort of thing, but I think you are too!

Stuff to think about, anyway   Wink  We await the results of the experiment  Shocked

Regards,

Steve
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #7 on: January 05, 2007, 03:35:51 PM »

You guys are great!!!! Cheesy
I have a modified Johnson 500 that uses 572b's as modmulators. They have gone soft and I cannot find any good-ole american made replacements and I WILL NOT USE the CHINESE version. SOOOOOOO I am going to replace them with 813's triode connected. They seem to be a good choice. Zero bias, no screen voltage and will like to 2000 volts the 500 will supply. I will drive them either with a backward connected Hammond outpoot tranny or a cathode follower. I now have a cathode follower driver that was built by Guy...W1FRM using 6L6's but not sure it will have the monkey swing for the 813's.
Anyway, that is the plan. I was worried that the 813's would not like the 500 mod iron but I think it will be ok.......what do u think?Huh

Bill

That sounds like a lot of drilling and blasting, Bill.  The 813s may be taller and will certainly require a different (larger) socket.

811as are still cheap - may not last as long as 572Bs but they wouldn't require any changes.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #8 on: January 05, 2007, 04:40:07 PM »

Throw a couple 811As in and keep looking why hack a nice rig. Even if they don't last as long so what how many sets will last a life time. 813s will need 10 volts at 10 amps so you will have to cob in a transformer.
The only thing the 572Bs did for you was provide more plate dissipation.

blow a little air across the 811As if it is a problem.
Logged
kc2ifr
Guest
« Reply #9 on: January 05, 2007, 07:19:15 PM »

OK...I guess I should explain a few things. I bought the 500 already modified....the power supply modulator has only 2 toobs.....the modulators. All other toobs have been either solid stated or removed. The VR toobs are gone. The 500 will NOT be used as a linear for SSB so the screen VR toobs are not needed. The modulators (811a's) were replaced with 572b's. All the low level audio stuff was removed. The 572b's were driven by 6v6's through a transformer connected to the grids. I removed all that stuff and used a backward connected Hammond tranny to drive the modulators. I have a QSC solid state amp that I use to drive the Hammond.  Guy...W1FRM then built me a cathode follower using 6L6's. This worked great untill the 572b's went soft. SO.........I want to install modulators that will last a long time and that are NOT made in China. If I had a shit load of good 572b's, I would be fine.......but I dont.  So if the 813's triode connected will fill the bill....I think that would be good.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #10 on: January 05, 2007, 10:33:06 PM »

go for it if it has been modified. 813 will have a lot more emission with double the heater power. Remember tubes are a bit wider and sockets a lot wider.
I've seen 813s done as GG linear amplifiers so you might get lucky tieing the grids but I bet it will need plenty of drive. I like Steve's Zener idea !
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #11 on: January 05, 2007, 10:53:45 PM »

Bill,

My 2 cents worth.

Start by throwing a pair of 811a in the hole. They're still pretty inexpensive. See how long they last.
They might do just fine, since voice is very much low duty cycle stuff.

It's a bit over their nominal B+ rating, but they don't mind being a little red in the face...

I'd seriously consider a pair of 810s over the 813s. The 813s are cheaper, but much more difficult to properly implement, since they probably don't strap into triode so simply. I'd kinda like to bootstrap the screens so they drive with the plate, but that's not triode operation... and the Z of the 813 is 2x that of the 810 for the same B+, and the 810 will be loafing along and have the lower Z on the plate.

Fils, 10v @4.5a.

If you don't need like 400watts of audio then consider some 838s or 805s... they'll do a little less than that, but are super simple to set up, zero bias.

Btw, the max screen 2 dissipation listed is 54ma... for the 813.

The usual method of making a pentode into a triode is to simply strap the screen to the plate via a current limiting resistor.  One possibility, although I don't know how it would work, is to make a divider set so that the current is limited and the voltage applied to the screen tracks the B+ (and so the signal) but is some fraction of the full swing...  one could bypass the top of the divider with an appropriate cap, so that the signal swing is almost full, but the DC is in the ratio... dunno how that might work out, but seems interesting. Also not sure how that might effect the output impedance vs. a straight triode strapping. (or the linearity)

I think if I had some 813s that I could afford to fry, I'd try the current limiting resistor and go for straight triode strapping and see how it flies... but I'd throw in some 811a and maybe a fan first, see how they stand up.

            _-_-WBear2GCR
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #12 on: January 07, 2007, 10:19:59 AM »

Plenty of hams have connected 813 grids together and used them as triodes. Tried and true. Tom Vu rig has plenty of customers including Bi-Polar Craig. The 813 will easily outlast the 572B. 572 b is only an 811 with a shot of testosterone. JV 500 mod iron matching will not be an issue.  Problem is whether your driver makes enough smoke.  Consult Guy Engineering on that. Linear R.F. and audio service use similar numbers all around so one can simply look at linear amp circuits for ideas.
813 will work and you already have some. OTOH, low mu triodes are preferred for the service in question. That is not what you get with the grids all tied together in an 813. The 810 is a better choice as mentioned. It can schwing more plate current with less grid current as with any low mu toob. Bias circuit will have to be changed though. 810 can be looked at as if it were 1/2 an 833. Dunno anyone who connected the screen grid to the plate on an 813. Somewhere out in space is the tale of Collins considering use of the 813 as a modulator in the KW-1. That tale would hold some insight.
    You can extract at least the same amount of power out of a pair of triode connected 813s into the same load as a pair of 572bees. BTW the used ones look like bees with dark and light coffee stripes. The B stands for beaten. 
Logged
The Slab Bacon
Member

Offline Offline

Posts: 3934



« Reply #13 on: January 07, 2007, 11:15:01 AM »

Like Dave said, 813s with both grids tied together, zero biass is tried and true. Many people have built variations of the K1JJ / Tesla 360 and they work fine. i was talking to John KC2FXE on 160 last nite and his sounded fine.

Generelly speaking tying the screen to the plate is generally used for something approaching class "A" operation. Tying both grids together and driving them puts you somewhere in the AB2 - class "B" area.
An outboard amplifier should have enuff scrote to drive it easily. A pair of triode connected 813s should have no trouble making somewhere around 500w of audio peaks. 813's are rugged, cheap and plentiful, not a whole lot of thought process here!!
                                                                   
                                                                  the Slab Bacon
Logged

"No is not an answer and failure is not an option!"
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #14 on: January 07, 2007, 11:44:11 AM »

Yes.. A pair of triode connected 813s will make 500 watts of audio easily.   It takes about 30 to 60 volts RMS  to  drive them as modulators.  I have a HB speech amp with a pair of 2A3 in the output.  This speech amp makes about 90 volts at full scrote and it easily drives an 813 modulator.  I also use another HB speech amp with pp 6L6s with neg feedback that  drives 813 modulators...   Any PA amp with a 70V output should drive them fully......
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #15 on: January 07, 2007, 08:04:03 PM »

http://cgi.ebay.com/2-Cetron-572B-Power-Amp-Tubes_W0QQitemZ180072417780QQihZ008QQcategoryZ48710QQrdZ1QQcmdZViewItem

572B/T160L   I've bought from this guy
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #16 on: January 08, 2007, 08:00:33 AM »

That sounds like a LOT less work than punching the chassis, doing a bunch of re-wiring, installing a new filament xformer, etc.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
K1KFI
Guest
« Reply #17 on: January 09, 2007, 01:34:44 AM »

So..what about a pair of 805's in the modulator...pure triodes, eh?
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.074 seconds with 19 queries.