Heathkit AT-1 Rebuild

[W9ZSL]

I just scored more AT-1 parts including chassis, switches, sockets, components and coils.  I have just about everything except the cabinet and hardware.  My power xfmr which is a slightly lower voltage fits perfectly and a Triad 10 Hy 90 MA choke is almost an exact fit.  I have replacement parts for the old ones.  I'll check the RF chokes but probably will replace them as well.  I'll probably rack-mount and will make some modifications especially simplifying the power supply filter with a pair of 10 MFD @ 500V rather than a mess of 8's in series.  I also want to try a 6V6 in place of the 6L6 for a lower output to drive an 813.

My question is about the chassis and coil mounting bracket.  I've seen some pix of restorations and those parts are shiny copper.  The stuff I have is rather dull.  Were these parts copper, copper plated or painted?

[N2DTS]

I do not understand why a tube with less plate dissipation will give less power output then one with more if run at the same voltages.

In order to lower the power output, would you not need to run less voltage or current?

[VE3AJM]

In this case with subbing a 6L6 with a 6V6 ,  the important difference is the cathode emission not plate dissipation. Cathode emission is much less with the 6V6.

Have a look at the filament ratings for each tube.

6V6  6.3v 0.45a
6L6  6.3v 0.9 a

Al VE3AJM

[N3GTE]

A lot Heath equipment had copper plated chassis back in the '50s. Have seen some AT-1's with rusty chassis (copper plated steel) You might want to have a look at Oct '55 QST good article on the AT-1

Terry N3GTE

[N2DTS]

It does not seem right to me to limit power by using a tube beyond its ratings.
I would think that cathode emission limitations is something to stay away from, and will not most tubes exceed their
plate dissipation ratings before running out of emission?

In this case with subbing a 6L6 with a 6V6 ,  the important difference is the cathode emission not plate dissipation. Cathode emission is much less with the 6V6.

Have a look at the filament ratings for each tube.

6V6  6.3v 0.45a
6L6  6.3v 0.9 a

Al VE3AJM

[W9ZSL]

I think it's probably copper-plated.  As for the PA, a 6L6 is no problem.  I have one.  I just don't want to overdrive the 813.

[VE3AJM]

I do not understand why a tube with less plate dissipation will give less power output then one with more if run at the same voltages.

This was the question that I was addressing. I assumed that when you stated "same voltages", that the maximum voltage and operating ratings of the 6V6 or a tube with less dissipation, would not be exceeded when subbing out the original 6L6 or higher dissipation tube. The answer is simple vacuum tube principles.

I've never owned an AT-1, but I would think that by making the screen voltage supply adjustable on the 6L6, you could dial in the output/grid drive required for the 813.

Al VE3AJM

[N2DTS]

So if you want to reduce the power of your KW-1 or 4-400 rig you can just put a 4-125 in and its good?

[VE3AJM]

I answered your original question about the 6L6 and 6V6 substitution, that I quoted in my last post. If thats not good enough for you or you don't get it, thats your problem.

You can answer your own rhetorical or red herring type questions. Now its a KW-1? I did address W9ZSLs issue with reducing the output from the 6L6 stage in the Heath AT-1. Thats what I would do.

Do you actually read the replies? Is it something in the water there??

If not, go to any vacuum tube handbook. You will learn something perhaps. Most likely not, given your recent replies.

Al VE3AJM

[KA2DZT]

To reduce power, make the screen voltage adjustable to the 6L6.  Adjusting the loading of the AT-1 into the 813 will also adjust drive.

Fred

[W9ZSL]

My power xfmr is 790 CT no-load while the AT-1 iron is around 800.  There won't be that much difference after all.  The screen resistor is 22K 2/watt and I can easily use an adjustable as the bleeder then set the tap accordingly.  The AT-1 bleeder is 2 x 47K @ 2 watts each in series. Sounds like a plan.

[N3GTE]

In their stock configuration you'll be lucky at 7-10wts out. The 6AG7 will do about as much as the 6L6 in this set.The final runs as a doubler on all band but 80mtrs which knocks the heck out of the efficiency. Like 30 percent. Heath sez 25wts input they don't mention the output power.
Don't think that the 6L6would very happy with 700-800v on the plate. I'd say 300-350v tops the the 6L6. You note that heath didn't use a 807 or 6146 that's why. At 300v you are looking at 10 or 15wts with an 807. By using the 6L6 heath didn't have to use a larger power transformer and a large voltage divider to supply the 6AG7 at 300v and an 807 at full HV. With some modification the 6L6 can act like the 807 at 300v and putout about 15-20wts

Terry N3GTE

[N2DTS]

I really did not understand your reply and still don't.
If I want less power output I reduce voltage or current, nor change the tube to something with less plate dissipation.

People want to reduce power output in various rigs like rangers, to drive amps, and just pulling out the 6146 and plugging in a 2E26 is not the way to go, reducing the plate voltage does work (running everything on the low voltage supply), or reducing the screen voltage (some power reduction).
Its ok to run a tube at well under its ratings at lower power, so I really do not understand why the smaller tube.

Maybe you can recommend some good book to read on tubes, I really do have a lot to learn.

And I thought this was a friendly place to talk about radio, designs, problems and so on, but you do not seem to enjoy it.
 

I answered your original question about the 6L6 and 6V6 substitution, that I quoted in my last post. If thats not good enough for you or you don't get it, thats your problem.

You can answer your own rhetorical or red herring type questions. Now its a KW-1? I did address W9ZSLs issue with reducing the output from the 6L6 stage in the Heath AT-1. Thats what I would do.

Do you actually read the replies? Is it something in the water there??

If not, go to any vacuum tube handbook. You will learn something perhaps. Most likely not, given your recent replies.

Al VE3AJM

[W9ZSL]

Plate voltage is listed as 400 VDC max according to Heath's AT-1 manual.  I'll be just under that.  I'll use the 6L6 and see how it goes.  My supply will put out around 375 VDC. The exciter will feed a single 813.  My only concern is over-driving it but it looks like I'll be right in the ballpark.  With 2KV on the 813 plate, drive is charted at 4.3 watts for class "C" phone.  I should be good to go.

Started buffing the chassis.  It is copper-plated.  I'll have to rack-mount everything.  I won't have to get many parts because the Jonque Boxe is providing basically everything I need.  Some parts from the original kit can be used also because everything seems to be in great shape and up to specs including the RF chokes. Slide switches will be replaced by toggles, pilot lights added along with separate meters for grid and plate.

Will keep you posted and include pix. Thanks!

73,
Mike

[N2DTS]

Its best to run the 813 deep into class C, and changing the grid leak resistor varies the power needed to drive the grid and its bias voltage.
On my pair (of 813's) I have a big wire wound pot as a grid leak resistor plus the fixed bias.
Increasing the resistance lowers the grid current, increases the bias voltage and increases the drive needed.
I think its 30ma for a pair? and its close to -400 volts I run it at I think, and it takes the full output of a 6146 to get there if adjusted that way.

I never had anything bad happen driving a grid deep into class C, but have had some nasty stuff happen to the signal with too little grid drive....

[W9ZSL]

Well, If I need more drive, I can use my Kenwood 440 which has adjustable output, a DX-20 or a DX-60.  Doing a custom rebuild of an AT-1 will get me back into construction.  It's been awhile so I'm out of practice.  Desoldering the dang parts is a real pain but the main stuff like the band switch is already pretty clean.  The chassis has some corrosion; nothing major.  I brassoed a spot and it should clean up pretty good.  I'm dragging the shop vac down stairs to suck up some mouse poo and layers of dust.  Like I said, "It's been awhile."

Cat is coming along to clean out the meeses.

[N2DTS]

On the plus side, modern parts are better and smaller then the old stuff.

I assume you have seen all these:
https://www.flickr.com/photos/wb5kcm/2338604147/in/photostream/

Very simple rig!

[DMOD]

The Heath AT-1 makes a good exciter circuit.


I think what people have been suggesting here is that it would be easier to control power using a variable screen grid voltage on the 6L6, rather than to try and control it using different tubes with different emission currents.

Below is a suggested circuit for controlling output power by varying the screen grid voltage as is done in many exciters.

Rsg replaces the 22k, etc.


Phil - AC0OB

[DMOD]

The Heath AT-1 makes a good exciter circuit.


I think what people have been suggesting here is that it would be easier to control power using a variable screen grid voltage on the 6L6, rather than to try and control it using different tubes with different emission currents.

Below is a suggested circuit for controlling output power by varying the screen grid voltage as is done in many exciters.

Rsg replaces the 22k, etc.

Phil - AC0OB

[W9ZSL]

I'm thinking of substituting an 80K adjustable in place of the 2 x 47K in series bleeder in the AT-1 power supply circuit. The slider will supply screen voltage in place of the 22K.  Rather than using 4 x 8 MFD, I have a pair of 10 MFD at 500V in the cap input filter.

A few other notes.  I read recently that the .001 caps in the primary of the xfmr are not necessary.  Heath calls for .001 micas for all the bypass caps.  I have a boat-load of disc ceramics which are new and 1/3 the size.  I assume they will work just fine.

I also scratch my head at the 2 caps in the 6AG7 grid...15 and 100 pf???  Most circuits are more like 47 and 220.  However, if this is what Heath used, I'm inclined to run with it.

Forget the grid/plate switching circuit.  I'm hard-wiring in two meters.  Still debating the two octals for VFO and Modulator.  I have a 3885 xtl for starters.  Baby steps Sparks; baby steps.

[N2DTS]

100 pf dc blocking cap going to the tank circuit seems small also, maybe because the final multiplies the frequency?
The entire thing seems like a bad design.
Be nice to have the screen voltage adjustable in the 6ag7 as well, drive must be all over the place...

[DMOD]

I also scratch my head at the 2 caps in the 6AG7 grid...15 and 100 pf???  Most circuits are more like 47 and 220.  However, if this is what Heath used, I'm inclined to run with it.

Heath wanted a certain amount of RF feedback to sustain oscillations and to keep the waveform clean, so I wouldn't change the values or the ratio in that capacitive voltage divider.

If those caps are temp compensated caps, I'd use the exact replacements.

I prefer ceramic capacitors of sufficient voltage ratings over dipped mica in most 1 to 30 MHz circuits.

Phil -AC0OB

[DMOD]

100 pf dc blocking cap going to the tank circuit seems small also, maybe because the final multiplies the frequency?
The entire thing seems like a bad design.
Be nice to have the screen voltage adjustable in the 6ag7 as well, drive must be all over the place...

Actually, the coupling capacitors are low value so as not to over-couple the oscillator to the 6L6. What you want is relative good isolation such that when the 6L6 impedance changes for different loading conditions, the oscillator frequency isn't "pulled."

I.E., loose coupling between stages guarantees the following stage impedance is not "reflected" back to the oscillator.

In addition, you want a constant, regulated screen voltage for frequency stability as shown below.

Phil - AC0OB

[N2DTS]

I have a similar circuit in use as the BFO in my homebrew receivers, xtal osc with a pot in the screen to set the level of the BFO signal, from off to maximum, 455 KHz xtal, and changing the screen voltage does not change the frequency at all, not even 10 Hz.

Now if it was a VFO it might, I never tried that...
My local oscillator circuit in the homebrew receiver looks very close to that same circuit, just with a coil and cap in place of the crystal.

[DMOD]

I have a similar circuit in use as the BFO in my homebrew receivers, xtal osc with a pot in the screen to set the level of the BFO signal, from off to maximum, 455 KHz xtal, and changing the screen voltage does not change the frequency at all, not even 10 Hz.

How do you measure any output at all when the screen voltage is below say 10 volts?

A 455 kHz crystal and holder probably has such a high capacitance that it swamps any tube or circuit parasitic capacitance.

Anytime you change screen voltage you change the amount of current flow, which changes the tube's dynamic capacitance.

For a high frequency oscillator, you want a constant tube capacitance so the overall circuit capacitance doesn't change the operating frequency, which is why you want a regulated screen and/or plate voltage.

Phil - AC0OB