813 Rig

[N8YX]

However MANY 813 rigs (that's a pair modulating a pair sometimes in push-pull, sometimes parallell) run 2200 or even up to 2500 volts on the plates and do 375 watts carrier for decades. (@ 100% modulation that 1500 watts pep, legal limit). External cooling HELPS (although many don't provide it given enough breathing room) and of course the power supply needs to be sufficient.

Would you care to provide a link to schematics and design notes of such, so they may be analyzed by any prospective constructors?

[KX5JT]

Would you care to provide a link to schematics and design notes of such, so they may be analyzed by any prospective constructors?

Okay... a link

http://www.amwindow.org/tech/htm/813/813.htm

Some people run up to 3K on the plates.

Here is an ongoing project based on aforementioned design:
http://home.comcast.net/~rsoennichsen/813Rig.htm

Note in the original schematic "Plate: Load 250ma to 500ma depending on power output desired"

[WØTKX]

The Tesla 300...

[N8YX]

Okay... a link

http://www.amwindow.org/tech/htm/813/813.htm

And in the bottom left corner we have...?

My original power figures.

"Some people" have also been known to place 5+ Kv on the anode of an 8877 and drive the thing with 120w, but it won't stay in viable shape for long.

ETA:

In the "ongoing" project page I see an EP of 2kv indicated on the meter. Obviously, someone is attempting to make them last longer than the weekend - and adhere to regulations while doing so.

[KX5JT]

You're simply wrong. 813 rigs running at 2000 volts and 400 mils are common and in use all the time.

I posted the schematic and design and a project in progress. There are many of these rigs already in use.

[N4VGB]

Would you care to provide a link to schematics and design notes of such, so they may be analyzed by any prospective constructors?

Fred, you don't really NEED to push the 813 over 2kvdc on the plates for max legal output, but some do it successfully.

WB4GWA/Ron does it this way: http://tinyurl.com/4dmckxp

K1JJ/Tom does it this way: http://www.amwindow.org/tech/htm/813/813.htm

KF4TXQ/SAM has one also: http://www.mindspring.com/~cacutts/r...sc/kf4txq.html the 813 rig is visible sitting high center left in the first pic with no side panels (813 RF finals clearly visible).

A pair of 813 RF finals is easily capable of 375 watts of carrier power. As modulators you can triode connect them as class B and apply 810 specs to the setup or as some do (WB4GWA), use them as tetrodes in class AB1. Plenty of audio power available either way.

On a road trip once I saw an old gents HB AM rig in which he'd triode connected his 813s in the RF final and modulator, applied 810 tube specs and was running it at about 250 watts of carrier 100% modulated. I wanna say he was running it in the neighborhood of 1.5kvdc or a little less? just from memory. Which makes sense because the most popular old 250 watt AM BC xmitters back then were 810X810 designs using similarly low B+ voltage.

The 813X813 and of course the single 813X811 setup must have been the 2 most popular HB AM rigs in the post WWII period. The availability of all those dirt cheap war surplus tubes would seem to make it a no brainer. Those also seem to be my most seen oldies HB AM xmitters in junk piles at hamfests over the years.

Anyway, once you are in the 250-375 watt range of carrier power AM there won't be enough difference in the top to bottom of that range for anyone on the receiving end to be able to perceive the difference.

[N8YX]

Well...I guess Art Collins had it all wrong when his firm designed the ART-13.

125w ICAS dissipation x2 tubes = 250w.

2kw PEP input x 0.75 (assume 75% efficiency at lower frequencies; this decreases as you go higher owing to relatively high inter-electrode capacitances) = 1500w PEP output. At the same time, you've exceeded the total rated ICAS plate dissipation by a factor of two.

Were you to pull this stunt with a solid-state PA you would almost certainly destroy the active devices. As it stands, you may get away with that kind of power level from a pair of 813s in Class C but unless some serious forced-air cooling is employed (along with a relatively short duty cycle) they aren't going to last.

I draw upon my own experiences with the tube in making these statements, and know how bright a pair of them will get at a mere 600w CW output when run in Class C. I certainly wouldn't want to run them at that power level for long.

[N8YX]

You're simply wrong. 813 rigs running at 2000 volts and 400 mils are common and in use all the time.

I posted the schematic and design and a project in progress. There are many of these rigs already in use.

Looking at the following spec sheet, one sees an ICAS parameter entry of 2000v max at 400ma max:

http://faculty.frostburg.edu/phys/la...datasheet2.jpg

One also sees a per-tube output of 300w max at those values. Even if we're talking PEP here that's still on the order of 1200w.

Increasing the plate voltage (and with it, dissipated plate power) far beyond the specified maximum falls outside RCA's design criteria for the tube and will result in shortened service life.

FYI, there are a lot of folks who claim 300w+ PEP output in the HF region from a pair of 2SC2879s but I certainly wouldn't want the thing on the air near me. On the other hand, their component suppliers probably do a rather tidy business in spares.

[N4VGB]

Well...I guess Art Collins had it all wrong when his firm designed the ART-13.

125w ICAS dissipation x2 tubes = 250w.

2kw PEP input x 0.75 (assume 75% efficiency at lower frequencies; this decreases as you go higher owing to relatively high inter-electrode capacitances) = 1500w PEP output. At the same time, you've exceeded the total rated ICAS plate dissipation by a factor of two.

Were you to pull this stunt with a solid-state PA you would almost certainly destroy the active devices. As it stands, you may get away with that kind of power level from a pair of 813s in Class C but unless some serious forced-air cooling is employed (along with a relatively short duty cycle) they aren't going to last.

I draw upon my own experiences with the tube in making these statements, and know how bright a pair of them will get at a mere 600w CW output when run in Class C. I certainly wouldn't want to run them at that power level for long.

Fred, the old ART-13 was designed with a lot of "head room" in the design for long life in severe military application. It's built in "battle ship" fashion (and don't I just love that "fashion") to survive the vibration and poundings it received in the B-17 airplane. Acording to the altitude of the plane it ran either 1kvdc or 1.4kvdc via an altimeter switch in the dynamotor setup on the 813 plate. Most hams using the ART-13 for AM work will slap 1.5kvdc on the 813 plate and run the sucker all day long in old buzzard AM transmissions. That puts the 811As in the modulator at their max rated voltage but none have ever reported a problem to me with doing so? Of course you don't drive the modulators anywhere near the max output in that setup.

Graphite plates will take abuse and survive much better than a tantalum plate. I don't know how many 833As I've seen with holes burned in the plates, it's a bunch. But the 833C graphite plate version will handle the same abuse and more with no complaint.

The max voltage and wattage rating of a lot of tubes were exceeded in many designs and applications. This setup comes to mind: http://bama.edebris.com/manuals/marmax/kw52
The first thing that kinda jumps out at you is the 5kvdc on the plates of those 4-250A P-P RF finals used in the design. 1kvdc higher than Eimac says is the max voltage rating and this design is from back when the FCC rule was 1KW DC input for amateurs running AM. It's one of those moments at first reading. But it quickly becomes clear that this is a controlled carrier setup that never exceeds any tube max parameters for more than a few milliseconds on voice peaks. Put processed audio into it and you'd meltdown those 4-250As in short order. The same is true of most max power AM 813X813 rigs, use processed audio and you're in trouble quickly.

An extreme example would be the old RCA 6293 pulse modulator for radar. It's absolutely nothing but a rebranded 6146B. Look at the specs for it in one of the old RCA manuals and you do the old a bunch of times. Until you come to the "duty cycle" number, very low. I used to scarf up every 6293 I saw at hamfests for change in my pockets while others were paying dollars for 6146B/Ws.

One of the things that may be in play here is that a lot of the old timers didn't believe in pushing the AM envelope to 100% positive. They'd set their average voice positive peaks at around 95%+ which usually would leave their negative peaks at 80-90%, realizing that on the receiving end fidelity is maximized in this fashion.

We could also delve into the age old debate over class C service? Defined as less than 180 degree conduction. How much less? Barely class C or deep into class C? If we go deep into class C service wouldn't that allow the max tube parameters to be fudged on a little also?

[KX5JT]

Art Collins designed the ART-13 to be very conservative and designed it to be run much lower than the tube lineup could handle. Remember, those were designed for airplanes during the WWII using dynamotors for the power supply. Airplanes at several thousand feet did not need much power to communicate effectively to their base stations. So he used overrated tubes so they would be rugged in the environment of airplane vibration altitude etc.

Remember N8YX, at the beginning of this thread you actually said

"Depends on how many 813s you use. A pair with ~2200v on the plates will realistically give you 500w PEP at the lower HF ranges. "

We were talking about class C plate modulation and the figures you gave were way off and very low. Even the conservative RCA specs say a pair in this service will give 300 watts of carrier power (1200 watts pep). Pushing those tubes just a little bit higher to get to 375 watts was TIME TESTED since the 1940's to be okay without shortening to the tubes lifespan significantly. It's been done over and over and over and over. It's not that far up from the original IACS specs friend.

Admit, you were mistaken with your first response and are now trying to save face by swearing up and down that the this is pushing the tubes WAY TOO FAR but it's not. You were thinking in terms of linear class AB1 service. It was apples. These are oranges.

Now, to the originator of this thread. I wish you the best success if you decide to take on the endeavor of building a classic 813 AM rig. Whether you run it conservatively at 300 watts carrier or 375 watts legal limit will not make any noticeable difference to the stations you are working. Running your rig at 300 watts might gain you a little more tube life and might be a good idea. We are amateurs. We experiment. We push the boundries. We sometimes blow things up. I understand you are not designing a commercial grade transmitter to bring to market. The small difference N8YX is harping on is not nearly the problem he is making it out to be.

Besides, you never even claimed to be making it a "legal limit" transmitter. You just want to build a classic 813 rig. I say go for it!

Have a great day everyone!
Peace, love and glowing tubes (even if the plates show a little color!)

John KX5JT