Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
3-to-4 parallel 807 HF Amp




 
The AM Forum
February 18, 2020, 01:41:47 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: 3-to-4 parallel 807 HF Amp  (Read 1583 times)
0 Members and 1 Guest are viewing this topic.
W4AMV
Member

Offline Offline

Posts: 567


« on: October 07, 2019, 09:49:05 AM »

Good day,

Prior to cutting metal, looking at a 3-to-4 // 807 HF amp, class C, CW.
VFO and Exciter are done, 5-10 W.

I have a surplus of good 807 and would like to go beyond the usual 2 tube affair.

Looking for a reference design or a sanity check moving to 3 maybe 4, beyond which I suspect the law of diminished returns.

All thoughts are appreciated. Attached is what I have in mind. Looking at the torpedo mount of the 807 or sub chassis mount No tube shields please. Try to keep in/out from talking from a decent layout. Have 1 kV at 500 ma. Single band for now, 40 meters.

I need to tie down the parasitic suppressor items. Apply HV and remove cutoff bias in Tx. VFO and Exciter keyed during Tx.

I should add, I estimate the Zin at grid 1 ~ 5 k and the total plate load at 800 ohms.
The input match may change to a simple L match plus a small pad.

I'll add grid and plate shunts for meter readings.

Tnx! Alan


* 4x807HF.jpg (45.63 KB, 995x546 - viewed 175 times.)
Logged
Carl WA1KPD
Contributing
Member
*
Offline Offline

Posts: 1548



« Reply #1 on: October 07, 2019, 02:54:11 PM »

Great project.
What is the exciter?
What do you expect for power out?
Carl
Logged

Carl

"Okay, gang are you ready to play radio? Are you ready to shuffle off the mortal coil of mediocrity? I am if you are." Shepherd
W4AMV
Member

Offline Offline

Posts: 567


« Reply #2 on: October 07, 2019, 03:07:48 PM »

Thanks Carl. Well hold on to your seat!

Homebrew VFO driving a newly restored Heathkit AT-1.
The AT-1 will drive this amp.

I am looking for best case 200 W out. I would expect 13 dB power gain from the 807's. At 5 W in, 100 W out.

However, with the AT-1, best case is probably 100 W as I may have to pad the interface between the AT-1 link coupled tank and the inputs to the 807 stack.

Currently the AT-1 is doing better than 10 W out on 40 M. So its a decent OLD exciter!

Alan
Logged
Carl WA1KPD
Contributing
Member
*
Offline Offline

Posts: 1548



« Reply #3 on: October 07, 2019, 03:31:36 PM »

Alan,
Well that is very cool and in line with one of my dream projects. So I will be interested in your final design.
I have a 2E26 exciter that would would work nicely with that.
Please keep me posted!
Carl
Logged

Carl

"Okay, gang are you ready to play radio? Are you ready to shuffle off the mortal coil of mediocrity? I am if you are." Shepherd
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2092


Making AM GREAT Again!


« Reply #4 on: October 07, 2019, 05:17:24 PM »

Push pull parallel?  Would be cool!

--Shane
KD6VXI
Logged
PA0NVD
Member

Offline Offline

Posts: 615


Nico and Chappie (Chappie is the dog...)


« Reply #5 on: October 07, 2019, 05:30:55 PM »

Must be a lot of fun to build an amp like that. The 807 still attracts me in many aspects. I think we all experimented with 807's. Please keep us informed. I will do something similar with 2 RS1003 tubes when the parts arrive and add cathode modulation. I do not have sufficient 807 tubes, otherwise I should make the same choice
Please keep us informed.
Logged
Detroit47
Member

Offline Offline

Posts: 490


N8QPC on your AM dial


« Reply #6 on: October 08, 2019, 10:44:31 AM »

I used to have an 8 tube Elkin amplifier . It ran 8 1625's push pull on 10-11 meters. It was a good design back in the days of analog TV it didn't touch channel 2. I have about twenty of these tubes still new. No one seems to have an interest in them.

John N8QPC
Logged

"We are all born ignorant, but one must work hard to remain stupid."  Benjamin Franklin
W4AMV
Member

Offline Offline

Posts: 567


« Reply #7 on: October 08, 2019, 11:27:40 AM »

Very nice John. So 8 push pull. Was that in pairs of 4 or in pairs of 2. That is to ask,  how were they power combined after each push pull stage. I can imagine the look of such an amplifier, awesome. Quite an affair!

And how was the push pull transformer designed? Link coupled?

Tnx, Alan



Logged
Detroit47
Member

Offline Offline

Posts: 490


N8QPC on your AM dial


« Reply #8 on: October 08, 2019, 12:05:37 PM »

I found this picture on another  web site. They ran in 4 parallel link  coupled. Not very pretty but it worked well.

Johnathan N8QPC


* 8tube16254_sm.jpg (275.75 KB, 644x483 - viewed 168 times.)
Logged

"We are all born ignorant, but one must work hard to remain stupid."  Benjamin Franklin
KC2ZFA
Member

Offline Offline

Posts: 380



« Reply #9 on: October 08, 2019, 01:23:49 PM »

there's a 3x1625 PA driven by one 1625 by W3BYM in a recent ER issue but I don't remember now the issue #.
Logged
w4bfs
W4 Beans For Supper
Contributing
Member
*
Offline Offline

Posts: 1316


more inpoot often yields more outpoot


« Reply #10 on: October 08, 2019, 02:54:25 PM »

a worthy project .... the trick to good performance and tube longevity is to get even load sharing .... working at Vp = 1kV is going to make a stable amp more difficult, especially considering the longish internal element leads .... you need individually variable bias for each tube and a way to meter each cathode current to even out each Ip .... 10W drive sounds just about right for 4 tubes .... be sure to neurtalize   Cool
Logged

Beefus

O would some power the gift give us
to see ourselves as others see us.
It would from many blunders free us.         Robert Burns
W4AMV
Member

Offline Offline

Posts: 567


« Reply #11 on: October 09, 2019, 10:49:17 AM »

Yes, load sharing and tube to tube instability is a concern, the so called odd mode that surfaces in SS amplifiers. I added series R's in the grid to grid and a little low freq degeneration in the cathode return. The input Z is forced to ~ 500 ohms with the 2.2 k R's return and the PI net input is low Q, about 4 to match to 50 ohms. Bias tweaks for each tube, to painful. If I need to add neutralized feedback I can. Don't want too and trying to obtain GG style operation with lower drive. Added meter circuits and next the PS.

Thanks, Alan


* 4x807HF_v2.jpg (55.97 KB, 974x570 - viewed 116 times.)
Logged
WB2EMS
Contributing
Member
*
Offline Offline

Posts: 630



« Reply #12 on: October 09, 2019, 02:30:14 PM »

I recall seeing an amp described in one of the mid 60ish ARRL handbooks that used a bunch of 1625's in parallel. I think the number might have been as high as 16 tubes. 1625's were cheap and plentiful. IIRC, the plate impedance of all those tubes in parallel was so low that the designer said a matching network wasn't really needed, just harmonic suppression. A ham I met in Rochester many moons ago had built one so I saw it in the flesh.

Googling around found this, but I don't  think that's the article I remember.
https://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/qst/1955/06/page21/index.html

I think it used a pair of husky TV transformers for the PS and only ran about 800V at substantial current. Kind of a brute force way to get a fair amount of power using commonly available parts. I'll have to look around in my old handbooks to see if I can find it. Must have been in QST at one point.





Logged

73 de Kevin, WB2EMS
KK4YY
Member

Offline Offline

Posts: 500


Your best isn't as good as you can be.


« Reply #13 on: October 09, 2019, 03:20:36 PM »

If you mount the 807 tube sockets below the chassis, consider leaving enough room to install 807W/5933 tubes. The 807W has a wider base but is an inch shorter than the 807 and has shorter internal lead lengths.

If 807's give you trouble, the 807W might be the cure. Wink


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
W4AMV
Member

Offline Offline

Posts: 567


« Reply #14 on: October 09, 2019, 04:11:49 PM »

Thanks! Good inputs!

A colleague around the corner caught a missing cap and also suggested bypassing each 2.2 k grid R. I am always hesitant on multiple bypass for fear of creating an anti-resonant case. Anyway, easier to take out then add!

QST had a nice paper on something similar, used 16 tubes, 6GJ5. Was QST Jan 64.
The amplifier is BB and looks very similar to the architecture of a DISTRIBUTIVE AMPLIFIER.

That input side as drawn is confusing. Sorry. Here it is by itself abbreviated.

Alan


* 4x807HF_v3.jpg (59.21 KB, 973x567 - viewed 112 times.)

* 807_input.jpg (34.6 KB, 854x455 - viewed 103 times.)
Logged
WZ1M
Member

Offline Offline

Posts: 375


« Reply #15 on: October 10, 2019, 03:01:05 AM »

807 = 6L6 with a plate cap.
Regards,
Gary
Logged
WBear2GCR
Member

Offline Offline

Posts: 3770


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


WWW
« Reply #16 on: October 15, 2019, 01:18:34 PM »

Gary,

True at lower plate voltages and lower freqs, especially audio freqs.
But in practice the 807 can be "pushed" a bit more than the 6L6...
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
W4AMV
Member

Offline Offline

Posts: 567


« Reply #17 on: October 15, 2019, 05:25:25 PM »

Yes, and furthermore there are differences even among the 6L6 series.

The 6L6-GB which may have been used in the Heathkit AT-1 I restored is rated at plate V of 360 V with plate dissipation of 19 W.

While the 6L6-GC is 500 V and 30 W dissipation. I think the GC maybe a bit closer to the standard issue 807.

I am going to try to push these 4//807's north of 750 V plate and Ip ~ 100 ma per tube.
Logged
SM6OID
Member

Offline Offline

Posts: 185



« Reply #18 on: October 21, 2019, 10:03:45 AM »

Hi!

Quite some time ago I made some experiments with 3x 807 in parallel configuration.
As far as I can recall, I had at least 1 kV under load on the plates and around 350 V on the screens.
This happened at the age of 16 hence the instrumentation was limited, actual output was not accurately measured but it was estimated to be close to 200 W give or take.
Screen dissipation was with in limits, that was checked...
Plate current meter was a 0-300 mA meter, it was pegged thoroughly...

Anyway, grid circuit was  swamped and amp was behaving ok.

Logged

RADIO: 51J-4, R-390A, SP-600 JX-21, BRT-400, Set No 19, T-47/ART-13, RF-590, SRT CR91, BC-312D, BC-348Q, HF-8020/8030/8010A/8090,  and much more...

ENGINE: Zvezda M50 F6L (V12), Rolls-Royce Meteor mk4B/2 (V12), Rolls-Royce B80 (inline 8 ) and much more
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 0.058 seconds with 19 queries.