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

donate to eham
   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Broadband Tube Amplifier  (Read 7928 times)
TANAKASAN
Member

Posts: 933




Ignore
« on: December 10, 2008, 04:12:06 AM »

If I want to build a solid state wideband HF preamplifier it's reasonably easy and there are lots of circuits from simple Class A stages to Norton 'noiseless feedback' circuits. How do I do the same job using a vacuum tube?

I have found a few schematics for RF tube amplifiers but all of them have a tuned circuit on the output (possibly for impedance matching). Do any of you have a schematic for a tube amplifier that can cover 1 MHz to 30 MHz without tuned circuits and do you have any suggestions for a suitable tube?

Tanakasan

P.S. Yeah, I know it isn't QRO but this is where the expertise is.
Logged
K6AER
Member

Posts: 5690




Ignore
« Reply #1 on: December 10, 2008, 08:07:15 AM »

Tube HF amplifiers generally run class AB1 and AB2 and have a tank circuits not only for matching but to add the flywheel effect to completing the conduction cycle. Most tube amplifiers are in a partial conduction cycle for efficiency and as a result you need a tank circuit to complete the conduction cycle and also to add harmonic filtering as well as impedance matching.

HF class A amplifiers are very inefficient and take lots of wattage even when RF is not being produced. The amplifier is in full conduction for all 360 degrees of the RF cycle.

You could produce a Class A amplifier that would be in full conduction cycle and with a push/pull  configuration it would be broadband but you will still need low pass filters to take care of the odd order products (harmonics).  With the high impedance of the tube plate output the push/pull impedance output transformer would be a bear to design for coverage from 1-30 MHz.

 Audio amplifiers used in music industry are class A but their output transformer  are iron and audio is less stringent with impedance matching.

The IMD of a class A amplifier is very good with proper designs being as high as 50 dB. A typical class AB1 or AB2 HF amplifier is at best about 35 dB in a good design. This is generally not a problem for most transceivers produce two tone IMD distortion of only 30 dB. Some of the class A output transceivers can produce a two tone IMD as good as 45 dB but they are rare and given the distortions in receiver design, band with limitations and overall band noise most hams would never be able to hear the difference.

Bottom line is most hams are happy with grounded grid triodes or grid driven tetrode designs because their efficiency and don’t mind tuning the output network for maximum efficiency.
Logged
AD4U
Member

Posts: 2538




Ignore
« Reply #2 on: December 10, 2008, 09:58:21 AM »

To the best of my knowledge the closest any manufacturer has ever come to what you described is either an automatic motorized auto tune amp or something like the Alpha 78 which had a sepatate pre-tuned tune and load capacitor for each band 80-10 meters.  

With the Alpha, all you did was select the desired band with the bandswitch and talk, assuming your antenna was a good match.

If a true broad banded tube amp was practical, somebody would have already done it.

Dick  AD4U
Logged
G3RZP
Member

Posts: 1225




Ignore
« Reply #3 on: December 10, 2008, 10:34:33 AM »

A broadband tube amp is best built with a distributed load amplifier. In that, you have a transmission line with the tube grids tapped along it, and similar with plates. See the 1965 edition of Millman and Taub, Pulse, Digital and Switching Waveforms, published by McGraw-Hill. Sec 5-8 p171 et seq.
Logged
K8AC
Member

Posts: 1897




Ignore
« Reply #4 on: December 10, 2008, 03:14:30 PM »

"If a true broad banded tube amp was practical, somebody would have already done it."

Indeed!  Central Electronics produced a broadband linear amplifier in the late 1950s to go with their 100V transmitter, which was also a broadband tube design.  The 100V used a pair of 6550 tubes in the final, not sure what the 600L amplifier used.  The 600L amplifier power level was only 500 watts input, but that was half the legal limit at the time.  I know that the 100V had a strange broadband inductor for each band, but never knew more than that.  Manuals and schematics are still available at the usual download sites.
Logged
G3RZP
Member

Posts: 1225




Ignore
« Reply #5 on: December 10, 2008, 11:44:35 PM »

Marconi introduced wide band high power (500 watt and 1kW) amps using the distributed amplifier technology back in the early 1960s. Fearsome beast with 16 selected 4CX250B in it for 1kW out. Covered something like 2 - 27.5MHz.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #6 on: December 14, 2008, 06:58:00 AM »

Don't confuse B and W's design with a real broadband amp.

The problem with a broadband amp is the impedance transformation. To have the necessary impedance transformation from a few thousand ohms down to 50 ohms, a matching network has to have a certain minimum operating Q.

B&W used a transformer that had a reactance compensated primary, but it had almost no harmonic suppression. They simply changed transformers to move to different frequency ranges.

If you used a B&W amp, you would have to follow it with a multisection low pass filter just like a transistor amp might use. Otherwise it fails FCC harmonic suppression requirements miserably.

Solid state amps are easier to make broad band because the impedances are very low. This makes it easy to design a transformer (which is what B&W used) without having to have multiple transformers, one for every major frequency change.

The reason no one builds anything like the B&W is the cost. You would have to switch transformers and switch low pass filters. The bandswitch cost alone would price it out of competition.

There are, as pointed out, amplifiers distributed along transmission lines but they are very limited in utility. This is because of the impedance limits of the transmission lines and the high plate impedance and stray capacitance of the tubes.

73 Tom

Logged
W8NF
Member

Posts: 64




Ignore
« Reply #7 on: December 16, 2008, 06:02:52 PM »

First off, to re-state Tanakan's original request - he is seeking info on PRE amplifiers.  Every reply so far seems to assume the goal is a power amp.

What you want is pure class A, true broadbandedness.

Such an amp is found in the late 60s and early 70s oscilloscope front ends, made by Tektronix and HP.  Find a manual for one of those scopes and you'll have your answer.

A diff pair made of a twin-triode (12Ax7) might be just what you want, or perhaps something broad enough could be made from a single triode with cathode degeneration resistor, output taken from the anode, and followed by a buffer (cathode follower).

See if you can find a service manual online for a Tektronix 453 scope.  That one, anyway, is a 50Mhz scope with tubes in the front end.  There were others, but I'm not as familiar with the model numbers.

Dave W8NF
Logged
KE4MOB
Member

Posts: 720


WWW

Ignore
« Reply #8 on: December 19, 2008, 12:54:38 PM »

"Marconi introduced wide band high power (500 watt and 1kW) amps using the distributed amplifier technology back in the early 1960s. Fearsome beast with 16 selected 4CX250B in it for 1kW out. Covered something like 2 - 27.5MHz."

I saw one of these on eBay about a year or so ago, remarkably over in G land...it was an awesome sight...lots and lots of glass.
Logged
K9MRD
Member

Posts: 384




Ignore
« Reply #9 on: December 22, 2008, 01:07:38 PM »

<<The 100V used a pair of 6550 tubes in the final, not sure what the 600L amplifier used>>

It's been a long time, but I am thinking the 600L used an 813.

Wayne
Logged
KH6AQ
Member

Posts: 7788




Ignore
« Reply #10 on: December 25, 2008, 03:20:01 PM »

Here is a simple untuned wideband preamp using a tube. A 6DJ8 dual triode. This was designed for VHF cascode amplifier service. The first triode half is configured as a grounded cathode amp. The second triode half is configured as a cathode follower for driving a 50 ohm load. Circuit gain is about 16 dB.

Triode #1 has a 200 ohm cathode resistor to GND. Shunt this with 0.01 uF. Connect the grid to GND thru any resistance. It can be 50 ohms to match the source. This is the amp input. The plate has a 1k resistor to a 100 to 150 VDC supply. Set the cathode resistor value to draw 20 mA of plate current. The gm of the 6DJ8 is 12,000 mhos at this current. This is quite high for a receiving triode. 1 volt swing on the grid gives 12 mA swing on the plate (for zero plate voltage swing). The voltage gain of this tube is 25 (for an infinite plate load).

Now for the cathode follower. Two 100 k ohm resistors provide a voltage divider from the plate supply to GND. Connect the midpoint to the triode #2 grid. From the cathode to GND place a 3.3k ohm resistor. This needs to be a 2 watt resistor. Connect the plate of triode #1 to the grid of triode #2 thru a 100 pF cap. Connect the cathode of triode #2 to the output using a 0.01 uF cap. This node can be connected to GND with a 100 k resistor to keep it from charging up. The amp gain rolls off below 1 MHz. To attenuate VHF gain you can place a 10 pF cap across the plate resistor.


Logged
VK4APD
Member

Posts: 1




Ignore
« Reply #11 on: July 06, 2019, 03:45:28 AM »

Look up transmission line transformers, as used in solid state amps. These will allow you to transform in ratios 4:1, 9:1, 16:1 etc.
Transform input up, and output down. The cable used must be less than 1/8th of a wavelength for the top frequency you are after. The lower frequency is limited by the inductance of the cable used, which you wrap around a ferrite toroid.
Use high Gm tube/s. Tubes used as pix amplifiers in TV will be needed, or TV tuner tubes e.g. triode 6GK5 or nuvistor 6DS4. They will need a rather low anode load resistor and peaking inductor, or an additional network between stages. Look up extending the HF response of tube amplifiers.
Logged
KM1H
Member

Posts: 5104




Ignore
« Reply #12 on: July 06, 2019, 02:13:43 PM »

Hmmmm, resurrecting an 11 year old post for what?

As far as the CE 600L it used a single 813 and was designed to be driven by the CE 20A or other 10-20W TX. My CE-100V has driven a NCL-2000 prototype since 1965 to 1200W.

A distributed amp I have here uses a multi stage SS amp followed by 13 RCA 8122 tetrodes (400W PD each) to run about 150W from 150 KHz to 150 mHz. Talk about inefficiency, I cant even lift the PS. However the price was right, the tubes are almost new and all matched so I cant complain as I use them for NCL-2000 service work.

Carl

Logged
G3RZP
Member

Posts: 1225




Ignore
« Reply #13 on: July 06, 2019, 03:51:17 PM »

The main use for such amplifiers was EMC immunity testing, where efficiency didn't really matter. The purely HF ones saw some use in Frequency hopping applications, both military and in ionosondes.
Logged
KM1H
Member

Posts: 5104




Ignore
« Reply #14 on: July 06, 2019, 05:48:03 PM »

Mine came from an EMC lab, didnt know about the other uses.
It was built by Amplifier Research Corp who is still in business at the same location in PA. aka ar Europe on your side of the pond.

https://www.arworld.us/
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!