4x813 amp

[WA1LGQ]

   Anyone out there using an amp with 4 813's in GG either as a linear of modulated amp? I have such a homebrew amp that was aquired from an esate and am finally ressurecting it. There is no documentation, and am not sure at this point exactly what plate voltage was used. I do have the power supply, but have'nt fired it up yet. It will be brought up slowly since I don't know yet the condition of the filter caps. Anyway, I was wondering what other parameters would be normal for an amp like this. Grid current, resting plate current etc. It has an adjustable bias supply that's connected to the cathode circuit. It has no tuned input circuit, I think it might have been used with a FT-101 which is not fixed at 50 ohms impedance output so it might be ok with drive from a tube rig.

        Thanks.......Larry

[K1JJ]

Hi Larry,

You might post some pics of it so we can drool... :-)

For grounded grid linear operstion, idle the 813's at about 10 ma each for 3KV or 15 ma each for 2KV. That's about 30W per tube.

The power output will depend on many things, so simply drive it until you find saturation and back off a little.  You probably shud add a simple L/C circuit at the cathode/fil.   Use a 3-4 section BC variable type cap  (1200pf?) and small coil to give you some decent Q at the low impedance. Q=3 is good.

I'm not sure what the combined screen/grid current will be in GG, but it's probably not important, cuz you can't really control it anyway. Just drive it for max output and watch the scope and you will be OK in linear service.  It will put out an easy 1500W pep with 2500V - and more with higher voltage.

Add some air if it's not already there..

Hope this helps, OM.

T

[WA1LGQ]

  Thanks for the info Tom. I found an amp on the web that has a rather broadband single coil input circuit that works for them, I may copy it. It would eliminate a band switch anyway. Here are the first pics, not too great, but you get the idea. The 4-813 amp deck is the one in the upper left of the cabinet and is now on a workbench, so I could supply better pics later. The power supply is below it in the cabinet, and in the pic you can see the final power supply on the bottom, and above it is another power supply panel that supplies the power for the modulator that is in the right side of the cabinet. There is another 813 deck in the right side of the cabinet that is 75/80M only and has only 2 813's in it. The wiring is a rats nest, so I thought the easiest thing to do first would be to get the power supply and linear 4x813 deck going first!

.......LB

[KL7OF]

What is the modulator?   Is that 4x813 deck  Class C with a screen supply?   Nice rig..TNX for sharing...Steve

[WA1LGQ]

There is a modulator in the right side of the cabinet Steve, but it is not for the 4x813 deck. Its for the 2x813 deck on that side. The modulator consists of 2x811's it looks like, although I actually have not looked at the numbers on the tubes yet. They look like 811's. There is a speech amp and/or driver section on another chassis. Have to remove each deck eventually and go through them. The 4x813 deck is a GG linear amp. I want to fire up the HV supply this week if I can get to it and see if it blows up or what. Should be fun........Larry

[K1JJ]

Looks like a lot of nice potential.

I think if it were my rigs, I'd clean them up and get the (4) 813 linear working first with as much voltage as you can make, up to 4KV.  Then work on the plate modulated pair X pair using about 2300V and see what it will do.

A nice cleaning and paint job wud do wonders for the rigs.


The GG amp looks like the coil may cover 40M, but not 75.   I wud also be careful using those doorknobs in the plate coupling, but they will probably be OK at that power level. Tough call.

It will be fun hearing you with a big signal coming outa E Htd, Larry.

T

[WA1LGQ]

Yeah Tom, the GG will be the first to get fired up, after the power supply is checked out of course. There is another section of coil that is not very visable back there. It has 160-20 on the band switch, and a vacuum variablethat you can see in there. The variable loading cap is pretty wimpy though, but its supplemented by some switchable fixed HV mica caps. The thing was on the air at one time, so I'm hoping for the best. So far found a few loose connections and loose hardware in the amp deck but for now the tubes light up and the bias supply and relays work. The filament voltage measured at 10.6 and there is a small variac on the filament transformer. There are no markings on the fil xfrmr at all.........Larry

[K1JJ]

Oh, yeah, I do see the second coil hidden near the front panel for the lower bands.  Both coils are pretty strapping.

That C2 loading can run out of range quickly. I found my 1500pf on 75M was not enuff for the big 4X1 rig cuz of the lower plate impedance. With four 813's you have, that's probably why he added the padders.  I've had a number of doorknobs blow up in C2 service.... :-)


Good on making progress and getting near to firing it up.  Let us know how the power output goes.  What is the max HV you can get for the linear if you short out the choke or do whatever you need to get it as high as possible?

Let me know when you get it on the air so I can give it a listen.

T

[W1AEX]

That looks like a nice piece of strap there Larry! I love the double-wide rack. Should look fine glowing and humming away in the living room! I definitely want to be listening on frequency when you bring that up on the air!

Rob

[w4bfs]

nice looking stuff ... noticed the 3ag type fuse holder on back panel in the hv lead ... this must go !!! ... the 3ag fuse cartridge is 250 V max rating and a blown 3ag fuse in a 2 or 3 kV circuit is almost a resistor and will continue to conduct from the redeposited metalization inside the glass/ceramic cylinder ...very very dangerous  because it may show open on dmm ohm test.  check with Tom about small wire open fuse which while somewhat problematic is better than something that could kill ...73...John

[WA1GFZ]

4CX3000A wants at least 2000pf on 75 to make the monkey happy

[WA1LGQ]

Ok, so after a bout with the flu or some other PIA virus its time to get back to the rig. Found that the variac was frozen in one spot. Had to tear it out give it to a friend who is good with such problems and he beat it, heated it froze it and sanded it back to life. Turns out that the Variac is a 120V unit wired into one leg of the 240v line. The power supply has a soft start system in there too. I found that there were bad solder joints on the bleeder resistor string, fixed that, and plugged it in. The transformer is a 230v pri, not split with 5200v CT secondary. Solid state rectifiers and a bank of electrolytics with a total of 48mf at 4500 working volts. The voltmeter on the panel reads 0-5kv. So anyway with the variac at 0 it runs at about 1200v no load. I did not want to crank it up much right away due to it not being turned on for many a year.  So its just sitting there with power applied and I can hear some light, muffled tapping or rapping sounds occasionally. Also the voltmeter needle fluctuates a little along with the noises. If I crank up the voltage a little, the frequency of the noise increases. I can feel a little vibration that corresponds with the noises if I put a finger on the transformer windings. It is an open frame transformer, so the windings are right out there, wrapped up of course. So what is that noise? Any clues?

...Thanks....Larry

[W3RSW]

wooo,  scary.

After a good visual inspection looking for carbon tracks in switches, etc, run the rig again with the lights out and look for external arcing in switches, terminal strips, bleeder resistors, insulators, etc.

If nothing external obvious, it sounds like some internal arcing somewhere. Either in the transformers or the filter caps.

Regardless, I think I'd take out the power transformer and bake it in the oven for several hours at, say some low setting under 180 degrees (get a decent thermometer for that.)  For that matter, bake all the transformers and chokes.

Hmmm, looking at the pix, I don't think I see any chokes, just two cap. input supplies. Let's hope the reason you got it is not because of a defective hv trans.-  But, hey, replaceable and look at all the neat parts you have.

Try to reform the caps after the transformers are baked. If that doesn't work, replace them.

Then if still arcing, start eliminating parts of the circuit, i.e. does the noise and meter wiggles still occur after eliminating the RF deck.   ..then.the last stage of the filter choke (if any) , cap. combo,.. then the first filter stage... then the rectifiers, etc. Last your looking at the HV trans. and variac combo.  Some hope required here..   

[N2DTS]

I could never bring myself to use anything that had electrolytics in it at those kinds of voltages!

Two electrolytics in series is as far as I go, and I wont even do that without having to for some reason.

Those things go and its going to be a huge mess plus the posability of injury if you are anyplace close to them.

Add the fact they are old electrolytics and I doubt I would even bother to try and use it.


Its oil filled for me, with at least 1000 volts under the ratings....

Brett

[WQ9E]

I would pull the tubes first to see if the noise continues; you could have some problems with one or more of the 813's.

This is probably a good time to check the equalizing resistors across the capacitor bank to make sure none of them have opened.

I definitely would not trust the insulation on exposed windings enough to touch them with my finger, be very cautious!

[WA1LGQ]

Yeah, I did not actually use a finger on the windings, but an extended finger, as in a plastic handled screwdriver between the thing and my fingers. Thanks for the thoughts. I think I will try to isolate various components, starting with the transformer. Maybe put a large resistor on the secondary. I should do the reform thing with the caps anyway, but I want to know right away if I need to look for another plate transformer. I thought about baking it, as it has been sitting for a long time. The rf deck was has not been connected yet, so its not coming from there. Maybe I should have an am radio on nearby and listen for arcing sounds.

.........Larry

[WA1QHQ]

Be careful when you fire it up Larry, Stan's ghost may come out of the rig and git ya!

[W1RKW]

Larry,
I had the same problem with my 813 rig which appeared out of no where one day.  I have the mod transformer and heising chokes insulated from the chassis.  I could hear the arcing in a receiver.  With only the HV applied and lights out I could see where the small arcs were occuring. The arcing was occurring on the choke and mod transformer frames to the chassis.  I provided better isolation and at the suggestion of JJ and GFZ also installed 1Mohm resistors to bleed off any residual HV on these components.  Instant cure for me at least. 

[N3DRB The Derb]

git rid of those old azz lytics!!

I bet most of the caps are dried out and are arcing inside the cans between layers.

[K1JJ]

Larry,

If an arcing is heard, but hard to locate after looking in the dark, then try using a rubber hose.  Make sure the rubber or plastic hose is well imsulated with no wire inside, etc.  Hold it to one ear and search around, just like a doctors steth.  I've found many arcs this way.

Also, sometimes by putting a large mirror behind the rig you can spot arcs more easily in the dark.

Is this rig from Stan, K1SJ in E.Htfd, or some other Stanley?

T

[WA1GFZ]

good chance one of the cap cans is flashing over. My first big supply i epoxy glued 200 uf/450 volt cans to a sheet of lexan. When it came apart years later the epoxy was burned between the cans of the first three caps.

[WU2D]

Don't forget the inpuke!

What are they doing on the cathode side? Switched individual Pi Input networks or a tapped tuned circuit approach?

Or is it the dreaded strappem all together and shove power in untuned?

Mike WU2D

[WA1LGQ]

It probably is bad old caps, but don't know yet. Yes Tom, the thing was once used by the infamous Stan, W1SJ of Black Sheep fame. It was probably built by him. Yeah Mark, maybe the sound is from Stan's ghost. Its probably haunted and won't run anything but ssb. No, thats not likely since in the other half of the cabinet there is another transmitter with its own modulator. Its a strange setup, from just a brief look it seems to be a single 813 driving 2 more 813's with a 2x811 modulator. I don't know yet what stage the modulator is connected to. There is also a separate hv power supply for the modulator. This transmitter is 80M only. The 4x813 amp is 160-20. Oh, and the cathode drive is the dreaded untuned coupling cap. More info to follow after my next session with the patient.......Larry
 

[KX5JT]

I'm far from an expert. However I had the very same symptoms emanate from my Heathkit SB-200.  Light sizzling sounds in small bursts that could also be heard in the reciever.  The High Voltage drops when they were heard, by a few hundred volts.  Dim front meter panel.  Eventually the intermittent nature gave way to steady failure accompanied by a slight plastic burning smell..  It turned out to be the original 43 year old electrolytic filter caps.  Visual inspection showed 2 or 3 of the bank of six to be swollen.  An upgrade to a new and improved Harbach powersupply board with beefier caps and a better overall design cured everything.  I run 15 watts of carrier into the amp and it has been happy since the upgrade.

KX5JT