Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Caascode Triode - Triode Receiver RF Amp  (Read 6798 times)
AA5WG
Member

Posts: 498




Ignore
« on: January 26, 2013, 01:01:59 PM »

Hi to all,

Has anyone built a triode cascode RF amp for the Drake R4C?

How hard is this to do?

Chuck
Logged
KB1GMX
Member

Posts: 786




Ignore
« Reply #1 on: January 26, 2013, 03:02:10 PM »

I've done it for SP600 using Nuvisters (very small triodes) and on a Tempo-one
for an experiment instead of the 6BZ6 I needed.  I consider it more a mechanical
task then anything as electrically its simple. In the case of the Tempo-one the
performance was par with the original tube (I used a pair of 6DS4 nuvisters
because I have a bunch).

There should be enough examples on the 'net. 

I'd be remiss that there is minimal advantage below 20mhz as there are many good pentodes
that offer excellent noise and gain performance to low VHF.  The one used in the R4C is very
good.  There are many other mods of greater value.  See the Sherwood Engineering site.


Allison
Logged
AC5UP
Member

Posts: 3900




Ignore
« Reply #2 on: January 26, 2013, 05:16:27 PM »

Popular tubes were the 6BK7 / 6BQ7 or 6J6 twin triodes often found as the RF amp in VHF TV and FM BCB tuners. If I was building one I'd likely go with a 6DJ8 because if the tube is good enough for Tektronix it's good enough for me.

Tek loved that tube.  Used a pantload of them in their 500 series 'scopes.

As for Nuvistors, they work well and last almost forever - but - the sockets are a bit small for point to point wiring and not that easy to find. If the Nuvistor concept is non-negotiable, look for junker Tek 500 series plug-ins at a Hamfest. If you can score something like a dual channel differential vertical amp you get four Nuvistors with sockets in two cush mounts with hardware.

BTW: Check the specs and you'll find maximum plate voltage on a Nuvistor is usually 70 vdc. They do not need a real HV transformer so a 24 volt jobbie can be wired with a voltage doubler to give you more than enough B+.
Logged

Never change a password on a Friday                
AA5WG
Member

Posts: 498




Ignore
« Reply #3 on: January 26, 2013, 07:11:05 PM »

Gentlemen,

Thank you for your information.  During the winter, here in northern Michigan, the noise levels go way down on certain nights.

As I listen to the R4C there is a ton of noise.  When switched to a 50 ohm dummy load this receiver still puts out a lot of noise.

I don't know if it mostly from the front end, the audio section, the old carbon resistors used, the way the tubes are heated or combination of all the above.

How was the performance changed with the Hammarlund SP600?

I could not find and examples on the net for the R4C.  Any suggestions?

Thank you,

Chuck
Logged
G3RZP
Member

Posts: 4731




Ignore
« Reply #4 on: January 27, 2013, 05:21:34 AM »

Is the noise from the antenna han 3 dB higher than the noise from the 50 ohm load? If so, you won't win anything.

Tek never needed AGC, so they could use the 6DJ8/E88CC with its sharp cut off. A better choice for a cascode is the ECC189/6ES8, which has a remote cut off characteristic. It's also not popular with audiophools, so is less expensive.

The 6J6 has a common cathode, so won't do on its own. Was fairly common as the gg stage after a 6AK5 in VHF converters, although a 6J4 is better.

IIRC, the R4C uses a 6EH7, and that has an enr around the 400 ohm mark: I doubt that you'll gain enough from a cascode to make the trouble worthwhile.
Logged
AC5UP
Member

Posts: 3900




Ignore
« Reply #5 on: January 27, 2013, 07:07:37 AM »

Yeah, I forgot the 6DJ8 has a sharp cutoff characteristic, but aside from that it's a damn fine piece of hollow state design.  If you look at the older Fisher FM tuners before they went Nuvistor in the 60's, I recall the 6ES8.  Coulda' been Sherwood, too.

The 6AK5 is a cute little booger but not a triode nor a dual section tube.  They're common in small signal RF amps, and if AA5WG wants to see how this was done in the R-1051/URR the squidmatic can be found here:  http://bama.edebris.com/download/military/r1051/R-1051.pdf

Excellent radio, but with the turret tuner a person might need to ignore some of the L/C filtering and switching upstream from the grid.

And all those transistors downstream................  Grin
Logged

Never change a password on a Friday                
WB6BYU
Member

Posts: 13342




Ignore
« Reply #6 on: January 27, 2013, 12:53:55 PM »

The Technical Topics column in the RSGB journal RadCom showed a plug-in
design to convert an RF amp to a cascode circuit by adding a dual-triode and some wiring
on an old tube base that plugged into the existing socket, with no required changes to the
rest of the radio.  I'd have to look around to find the circuit, but something like that might
be a relatively easy conversion that would be easily removed if desired.
Logged
KB1GMX
Member

Posts: 786




Ignore
« Reply #7 on: January 27, 2013, 03:40:19 PM »

Look here, first hit using google with Nuvistaplug..

www.hammarlund.org/files/NUVISTAPLUG.pdf 

It was an almost direct plugin for the 6BA6... read instructions as in the sp600
some touchup tuning was required and I'd expect same for the R4C.

It was an adaptor made by Raytronics to replace the 6BA6 used in the
Sp600.  IT was part snake oil and part real.  At 40M it would make no
useful differnce as propagated noise was still high enough.  at 10M and 6M
it did make a difference as manmade and propagated noise was lower and
the 6BA6 was not as good as some other later tubes (SP600 is older than
R4 series).

In the low VHF the choice of tube use was a big factor as gain and noise
tended to close in (gain going down but not noise) on itself.  In the 60s
some better tubes (pentodes) emerged 6BZ6, and others like the 6AK5
(uhf pentode) were used and helped greatly.   Triodes are generally
quieter and the The Nuvister being a triode was not that popular as
triode needed neutralization to be stable or used in grounded grid
format with low gain.  The cascode circuit was the known solution
and some of the better twin triodes were popular there and did much
better.  The Nuvister was applied that way and worked well but cost
made it more suitable for the demanding TV and FM environments
and VHF radios. 


Allison
Logged
G3RZP
Member

Posts: 4731




Ignore
« Reply #8 on: January 28, 2013, 01:11:28 AM »

The 6BA6 is the miniature version of the 6SG7. The 6AK5 was WW2 tube for low noise radar IF strips, and the famous Wallman cascode used a triode connected 6AK5 followed by a 6J4, a high transconductance (12000 micromhos) triode. That had a very small grid to cathode spacing to get the transconductance, and to minimise grid emission from sputtering of cathode material onto the grid, had a gold flashed grid. The idea of the cascode was  that the low plate load resistance on the first stage caused by the low input resistance (1/gm) of the second stage minimised the plate voltage swing of the first stage. This meant that Miller effect didn't reduce the input resistance of the first stage, while giving the lower noise of a triode.

Now this is all very well, but you need a triode with a high gm to get low noise in the first stage. To beat a 6EH7/EF183, that needs a gm of 5100 micromhos or more - not impossible, but the next question is 'What is the dominant noise source?' It may well not be the enr of the tube itself, but the Rd of the tuned circuit.

So before starting, measure the difference in output between the antenna connected and the load connected. If the antenna gives 3dB or more noise than the load, it  isn't worth changing. Switch the AGC off when making the measurement, of course.
Logged
KB1GMX
Member

Posts: 786




Ignore
« Reply #9 on: January 28, 2013, 07:19:49 PM »

That is why the 6DS4 Nuvister was desirable it has a mu of over 12,000.


Allison
Logged
WB6BYU
Member

Posts: 13342




Ignore
« Reply #10 on: January 28, 2013, 08:50:55 PM »

I did manage to track down the circuit I remembered in Amateur Radio Techniques
it used a 6BZ7 as a cascode plug-in replacement for a 6SK7 with the addition of 3 resistors
and 2 capacitors.

Might not be quite as good, but a similar approach could be used with other tube combinations.
Logged
G3RZP
Member

Posts: 4731




Ignore
« Reply #11 on: January 29, 2013, 03:43:21 AM »

Still comes back to 'is it worth it?' And 'If it ain't broke, don't fix it!'
Logged
AA5WG
Member

Posts: 498




Ignore
« Reply #12 on: January 29, 2013, 06:53:17 AM »

Sure it is worth it.  During the winter months there are days when the band noise seems to disappear to almost nothing.
I bet you could hear DX with a tin can and a string.

Most days this is not the case. 

Allison, the link you provided did not work.  Would you double check the link please?

I remember reading about how one should not tamp down the first stage of the radio, specifically the RF amp.

The R4C RF amp is tamped down at least 20 db or more.  The older tube handbooks say to remove or turn off the agc on the rf amp.  Move the agc down the chain and let the rf amp work full out.  That is, full out with in reason, i.e. grid bias adjusted correctly, ref. page 5.3 RSGB Radio Communications Handbook, 5th ed.  Also on page 5.2 - 5.3 of this reference is an explanation on how to reduce noise caused by degeneration. 

What are the best or top 3 tubes/valves for low noise cascode work?

Chuck.
Logged
G3RZP
Member

Posts: 4731




Ignore
« Reply #13 on: January 29, 2013, 08:55:00 AM »

What bands are we talking about? The lowest noise spot in the world is supposed to be South Georgia, where the noise is apparently about 10 dB below the ITU quiet rural median level. Even there, on HF, a noise figure of 5dB is all that you can use. How far above the rx noise, when on dummy load with AGC off, is this very low noise floor?

The R4C, according to the handbook, has a noise figure that equates to 11dB - pretty good for a 6BA6. I take it that you have checked that the sensitivity is up to the handbook level, which is  0.25 microvolts for 10dB (S+N)/N?

The simplest change to increase RF gain and lower noise would be to plug in a 6AK5 in the V1 socket - that would require no wiring or component changes - and see how much improvement there is. I know on the old HQ170A, using a 6AK5 with pin 2 clipped  off made a big difference on 10 metres. A possibly better tube substitution worth trying would be a 6BZ6 in the V1 RF amplifier socket, and again, you wouldn't need to change any components. Having said that, some experimentation with removing or changing the value of R3 (currently 15kohm) from the screen grid pin 6 to ground to increase the screen volts might win something with the 6BZ6.

Going to a cascode is more complex, as the 6ES8 (best choice) needs a Noval 9 pin miniature socket. Changing out the socket will be a bitch, as it's hidden under the bandswitch, plus you'd need something like a Greenlee punch to make the hole bigger.

The other alternative would still be pretty difficult in terms of mechanics and that's to make a small unit with a 6DS4/6CW4 nuvistor cascode (chosen so you can get the size down) and piggy back that in where the 6BA6 goes.

It would be interesting to know the conversion gain and noise of the 6EJ7 first mixer. Rough calculation suggests something of the order of 1500 ohms enr and probably 3 to 4000 micromhos of conversion conductance. The unknown is the plate load impedance, determined by the turns ratio of T5. There might be a chance of getting some improvement there, but that's a problem for another day......But note that getting the noise figure down may make IMD problems worse.

Perhaps an easier approach would be an outboard preselector with a 6ES8, as in pre 1960s practice, that you can switch out when not needed.

Hope this helps.

73

Peter G3RZP
Logged
AA5WG
Member

Posts: 498




Ignore
« Reply #14 on: January 29, 2013, 01:37:23 PM »

Hi Peter - G3RZP,

I admire your technical skill when it come to tubes and receivers.  If you have time could you share how you come up with the voltage numbers and gain numbers and more?

Thank you,

Chuck - AA5WG
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!