Date   
Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

ajparent1/KB1GMX <kb1gmx@...>
 

Gordon,

Yes don't.  Those vacuum fats have plat to grid capcitance and if not neutralized
(or even if it is) it will take off and get unhappy.

If anything lower might be the thing.  The key is you need enough RF grid voltage
to drive the tube but likley less than the total bias needed to get plate current.
Some tubes gt really unhappy or terribly non linear if they see grid current flow.

Grounded grid has some of the same issues but offers a naturally lower input
impedance and a bit less gain.

Allison

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Gordon Gibby
 

Thanks very much for sharing that schematic!   Even with the resistive loading it looked like the grid given PA of heathkit transceiver was doing about 16 dB gain.  I wasn’t brave enough to try it higher than 20 m 

Do you have any advice on whether we should try resistive loading higher impedance than 2500 ohms???

By using the tuned plate circuit of the 6cl6 driver, I guess we are already dealing with any grid capacitance in the process

I never did anything like this before (since I home brewed a 3x 811 a amplifier in high school ) it’s pretty cool for me to get back to this


On Nov 10, 2019, at 11:59, ajparent1/KB1GMX <kb1gmx@...> wrote:

The attached image was the solution of the 60 and 70s for hams (and CBers)
for 100W amp. it was cheap 10-13DB of gain.  Alas sweep tubes got expensive 
as people burnt them all up pushing those.  There were generations of them with
up to 6 (SIX!) for a KW input peak.  Not all were clean.

Some did grid driven and they worked save for over drive the grids and the
tubes protest by going away.  At 25$ each that is a bad thing.

For laughs a decade ago I took a 4CX250  and ran it grounded grid like that
and it was good for about 80W (on 6M) out in a simple circuit without forced
air cooling. The tube with screen bias (its a ceramic metal power tetrode)
added did an easy 300w out at 2000V (a very lethal power supply).  The
amp was terminally simple save for a typical tube amp tapped load coil
Pi network for multi-bands.  There are a long list of tubes that can can
be used for that that..

Allison
<courier_bl100_sch.pdf>

Re: Power output tests at 24v PA

Chris [N7CCX]
 

@mIguel, it is not designed to do 160, this was just a test.

7/10 ill check..


@mark variable multi meter

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

John Faivre
 

I have a HW104 with solid state finals and was thinking of doing the same thing. 
--
John Faivre WA9SGD

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

ajparent1/KB1GMX <kb1gmx@...>
 

The attached image was the solution of the 60 and 70s for hams (and CBers)
for 100W amp. it was cheap 10-13DB of gain.  Alas sweep tubes got expensive 
as people burnt them all up pushing those.  There were generations of them with
up to 6 (SIX!) for a KW input peak.  Not all were clean.

Some did grid driven and they worked save for over drive the grids and the
tubes protest by going away.  At 25$ each that is a bad thing.

For laughs a decade ago I took a 4CX250  and ran it grounded grid like that
and it was good for about 80W (on 6M) out in a simple circuit without forced
air cooling. The tube with screen bias (its a ceramic metal power tetrode)
added did an easy 300w out at 2000V (a very lethal power supply).  The
amp was terminally simple save for a typical tube amp tapped load coil
Pi network for multi-bands.  There are a long list of tubes that can can
be used for that that..

Allison

Re: Power output tests at 24v PA

ajparent1/KB1GMX <kb1gmx@...>
 

A note on 24V operating...

If you do that there will only be a bit more power is the result.
Since the gain of the finals does not increase with greater
applied voltage you still need more drive to take advantage
of it.  It only enables it to produce more power.

That goes back the the initial problem, not enough drive on
the higher bands from the decaying gain from the 3904s.

As to dummy loads and detectors,  Diodes are terrible load
and when connected to dummy load they change the
effective impedance presented back to the radio.
They also have a hard to predict offset (threshold)
that at best is approximated and untested.  Above a
few watts its not a big deal unless some one can show
a significant difference for 15V vs 15.5V (peak)
across a 50 ohm load. 

Then we get into peak vs p-p and RMS power.
Note most diode detectors are usually PEAK so
the actual is .707(give or take considering diode
offset) times the peak voltage (correct only for
sine wave). 

Also detected waveform makes a difference and assuming
a sine wave is not always valid.

Its those details that impact read power and why many people
get sometimes wildly different numbers. 

the simplest test is pit a fresh 1.5V battery across hte dummy load.
That should cause 30mA to flow and the diode detector if perfect
(and DC connected) will read 1.56V (fresh alkaline cell).  Likely
it will be a diode drop lower and that depends on the diode and
meter load presented to the diode.  Typical for 1n34 is about
1.45V,  1n5711 Schottky maybe 1.35V, and 1n4148 silicon
about .96.   FYI the resistor in the load if 50 ohms is dissipating
.045W (45 milliwatts) DC.  If it was 1.56V AC P-P its a lot less
(about 6mW).

Allison

Re: Power output tests at 24v PA

 

Hi Chris,

In your first column of power calculations, the formula is correct but you must have made a mistake in the spreadsheet calculation.  Either that or you didn't take into account the the probe gives Vpeak not Vrms so it must be converted.

This is shown correctly in the second column formula:  your reading squared divided by 100.  In converting Vpeak to Vrms you divide by the square root of 2.  However, since this is subsequently being squared, you can just put that in the divisor with the 50 and that makes it 100.

Of course both voltage measurements should have the diode voltage drop correction added back in before and any calculations are done.

Be aware that dummy loads/detectors can be built in different configurations so the formulas may vary (e.g. Elecraft's DL1 where the probe is measuring across only half the load).

73,


Mark

PS:  What is a VMM?

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Gordon Gibby
 

Fantastic!   part of the magic of ham radio is experimenting and developing more personal knowledge of radio and electronics.   

On Sun, Nov 10, 2019 at 8:38 AM Bill Cromwell <wrcromwell@...> wrote:
Hi,

In the 80s I built boots for an Ten Tec Argonaut 509 using the carcass
of a Heathkit DX-35. There is no receive circuitry involved. I biased
the single 6146 for class AB1 operation. I built a solid state T-R
antenna switch to keep the QSK feature of the 509. It served me for a
several years and then the 509 moved to a new home. The remains of the
DX-35 also went to another ham.

I have the CW portion of a Heathkit DX-100 down for overhaul. The AM
phone parts were all gone when I got the DX-100. I am going to dpo the
mods to include the linear operation in phone mode to amplify SSB
signals and keep the class C operation for it's CW operation - ala the
Apache (TX-1). Grid block keying has already been part of this
particular DX-100.

Sometimes QRP is just not enough.

73,

Bill  KU8H

On 11/10/19 7:35 AM, Don Richards wrote:
> Hi Gordon
> I ran across this a while back but never tried it.
>
> https://www.google.com/url?sa=t&source=web&rct=j&url=https://frrl.files.wordpress.com/2008/09/hotwaterforthek2.pdf&ved=2ahUKEwjwzsTQ09_lAhWsmuAKHQ1MCyYQFjACegQIARAB&usg=AOvVaw1PdjqVxSUEw-wayoqwuZyO
>
> 73 Donve3ids

--
bark less - wag more



Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Bill Cromwell
 

Hi,

In the 80s I built boots for an Ten Tec Argonaut 509 using the carcass of a Heathkit DX-35. There is no receive circuitry involved. I biased the single 6146 for class AB1 operation. I built a solid state T-R antenna switch to keep the QSK feature of the 509. It served me for a several years and then the 509 moved to a new home. The remains of the DX-35 also went to another ham.

I have the CW portion of a Heathkit DX-100 down for overhaul. The AM phone parts were all gone when I got the DX-100. I am going to dpo the mods to include the linear operation in phone mode to amplify SSB signals and keep the class C operation for it's CW operation - ala the Apache (TX-1). Grid block keying has already been part of this particular DX-100.

Sometimes QRP is just not enough.

73,

Bill KU8H

--
bark less - wag more

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Gordon Gibby
 

FABULOUS!!!!!!   That is way ahead of me, and probably an even better way to do exactly what I was doing!!!!!   Thank you for providinig that information!   

It looks like this works quite well.   I saw that in the schematic but buzzed past, never realzing that was probably the exact type step-up transformer just like what I was using!!!    

An even easier way to do what I was working on.    I'll be talking about these amplifiers to our local ham club on Wednesday.   What a great addition, your article makes.    

To resurrect these ancient and great transceivers, generally all one has to do is replace their electrolytic compacitors in the power supply.   And there are two on the main transceiver board as well.    Just that simple work (I'm going to make a printed circuit board for myself to make it easier) gives you what may be a very cheap power amplifier for the wonderful bitx devices.   

I've been able amazingly to get the transceiver for $100 and the power supply for $100 also, often.    What a huge amount of radio for such little $$.    Replace the VFO with a signal from the Raduino (it works great) and you have a computer controllable radio.    I've written that up before.   

Gordon

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Don VE3IDS
 

Re: Power output tests at 24v PA

Miguel Angelo Bartie
 

Hi Chris

I usually measure RF power with a diode and metal film resistors.
The diode barrier voltage is important for low power (below 2W).
The graph you get shows problems at 1.6, 7 and 10 MHz, in my view is that you are having problems with the low pass filter that has a lower than the expected frequency cut, thus reducing the power in these bands.
If possible change the value of the inductors downwards (by taking a loop, each inductor, for example).
After this measure the power again

73 from py2ohh miguel

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Rajendran Mayilsamy
 

Thanks Gorden for info and do post the rough schematic ..tnx n 73 Rajendran

On Sun, Nov 10, 2019 at 2:26 PM Gordon Gibby <docvacuumtubes@...> wrote:
What Allison suggested was to resistively load the transformer upping the impedance of the signal.   So I provided a 2500 ohm load on the hi impedance side.  

I’ll draw up a rough schematic later 

Thanks

Gordon




On Nov 10, 2019, at 03:44, Rajendran Mayilsamy <mrajsuba@...> wrote:

Dear OM Gordon Gibby, pl share the info shared by OM Allision about the using of Heath tube type txcvr final  and me to having a HW101 and would like to modify it..

Allison gave some much needed advice on this crazy idea months back,--This info pse


 73 de Rajendran,VU2SMM

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Gordon Gibby
 

What Allison suggested was to resistively load the transformer upping the impedance of the signal.   So I provided a 2500 ohm load on the hi impedance side.  

I’ll draw up a rough schematic later 

Thanks

Gordon




On Nov 10, 2019, at 03:44, Rajendran Mayilsamy <mrajsuba@...> wrote:

Dear OM Gordon Gibby, pl share the info shared by OM Allision about the using of Heath tube type txcvr final  and me to having a HW101 and would like to modify it..

Allison gave some much needed advice on this crazy idea months back,--This info pse


 73 de Rajendran,VU2SMM

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Rajendran Mayilsamy
 

Dear OM Gordon Gibby, pl share the info shared by OM Allision about the using of Heath tube type txcvr final  and me to having a HW101 and would like to modify it..

Allison gave some much needed advice on this crazy idea months back,--This info pse


 73 de Rajendran,VU2SMM

Re: Power output tests at 24v PA

Chris [N7CCX]
 

So my pemdas order of operations is wrong! 


I was wondering why I was pulling 16w on a ft60 and my bitx was pushing 70

Re: Using Vacuum Tube heathkit as Linear Amplifier for QRP

Gordon Gibby
 

(The little perfboard circuit in the right foreground is unrelated -- a circuit to deal with transmit/receive problems.....ignore)
The 49:1 balun (really, a Un-Un)  is in the blue electrical box with the banana plug outputs.  

Using Vacuum Tube heathkit as Linear Amplifier for QRP

Gordon Gibby
 



Finally got a chance to try my idea of using a heathkit ssb rig as a power amplifier for a QRP rig.   I've seen "amplifiers" for these rigs going for many hundreds of $$$.  I happen to be swimming in ancient heathkits with 2x6146 linear finals and tuned plate circuits on the driver as well (all good for getting rid of unwanted harmonics.....)

Allison gave some much needed advice on this crazy idea months back, and I finally got the chance to try it:

1 watt output from an exciter [I used an icom 718 that was readily availalbe] -- my only "power measurements" are ancient SWR meters with built in diode detectors so these measurements may be fairly INaccurate.

coax over to a 49:1 homebrew balun our group made for matching resonant half-wave antennas:   ours is a simple autotuner with 14 turns total of #18 teflon wire around a FT-240-43 (or FT-140-43) tapped at turn 2 to give us 7:1 turns ratio, 49:1 impedance ratio, and that checked out roughly with an antenna analyzer and some resistors

I loaded the output with 2500 ohms (temporarily made out of 4 10K resistors in parallel because that is all i had)
output of that via 6 inches of small coax to a 0.01 1kv ceramic capacitor connecting to the PLATE circuit of the removed 6CL6 driver, and to a nearby ground.

Output tested only with very brief bursts of CW -- on 80 meters and 20 meters, got approx 40-50 watts output measured also with an ancient swr meter/wattmeter into a 50 ohm load, everything on the Heathkit peaked for output.   Plate current rose appropriately.   Tuning circuit all appeared appropriately.

No obvious signs of any instability in this pilot test.    
Did not make any measurements with spectrum analyzer.   (In due time......)

That heathkit has other problems elsewhere but the finals appeared in fine shape.   Earlier VERY CRUDE measuresments with a siglent oscilloscope and some math suggested that the drive required at the INPUT to the 6CL6 stage would be less than 100 milliwatts, and that the drive required at the output of 6CL6 would be in the single digit watts range.    The above test appears to confirm those crude measurements.   Since it appeared the two 6146's could be driven adequately with single digit watts I opted to temporarily just remove the 6CL6 driver and connect to its (DC powered!!!!!!!  High voltage!)  plate circucit with a 1kv .01 blocking capacitor.   

Quite pleased with this very very initial test.   
At some point I'll repeat with a uBitx, but that's all for right now.
Gordon Gibby

Re: Power output tests at 24v PA

Jerry Gaffke
 

That's correct, assuming a forward voltage of 0.4 Volts across the diode.
According to fig 7 of the datasheet, that happens at a current of 20 uA when the temp is 25 C.
   https://media.digikey.com/pdf/Data%20Sheets/Fairchild%20PDFs/BAV21.pdf

Dividing the peak voltage (what you measure with a diode detector) by 1.414 
converts it to an RMS voltage, assuming we are dealing with a sine wave.
We must use that RMS voltage when calculating power.

Jerry


On Sat, Nov 9, 2019 at 12:42 PM, Mike Kilpatrick wrote:
I could be way wrong but I measure DC volts, add 0.4V then divided by 1.414.
Multiply the result by itself (square it) and divide that result by the 50 ohms load. Result is power in watts. 

 
 

Re: Power output tests at 24v PA

Mike Kilpatrick <mike.kilpatrick@...>
 

I could be way wrong but I measure DC volts, add 0.4V then divided by 1.414.
Multiply the result by itself (square it) and divide that result by the 50 ohms load. Result is power in watts. 


On 10 Nov 2019, at 04:56, Chris [N7CCX] <chriscoel@...> wrote:

Howdy all, I did last night some key down test (keyed the radio with a straight key) into a dummy load and did some power measurements..

I am seeing a bit higher than expected power outputs on 12V and then also 24! I took a single measurement on areas on the bands and then keyed down for a couple seconds and took the average VMM reading, I then took two formulas. One was QRPGuys Dummy load formula (which my dummy load is based on - Except the diode is swapped out for a BAV21 instead of a 72v diode). Other is K4EAA and even Elecrafts Calculations..

My results:
<dummyfile.0.part>




The graph is with the QRPGuys calculations which the formula is shown above.. My question is are these even close or accurate? The forumlas used above base on the exact builds they did. K4EAA uses 20 resistors to achieve a dummy load, meanwhile QRPGuys uses 4 resistors then they originally have a smaller voltage diode... Does someone have a better formula I should use? Signal is dumping into 4 3W 200OHM Resistors, cooled by a fan (cause they def are not rated for 20+ W :) then they feed into a BAV21 Diode with a .01UF capacitor on there to help smooth it out..


Either way the IRF510s didnt seem too upset from the extra power, but I am running a fan on one of the cases Sunil sells and its held up great (minus some fan noise i need to fix). I also need to run a probe on the 510s, since they are still on default heatsinks :)

Any help on confirming a better formula or even if I am accurate with my measurements helps!

Thanks