813 Rig - Slight change of plans...

[N8UH]

Well, my search for mod iron has come to a end, and I may have ended up with more than I bargained for!

Dahl transformer and reactor from a 5KW rig, 4-1000 modulator deck, Audio amp section, Meters - oh my! The seller is also throwing a 2500V plate transformer from a BC1H (2500V 1.5A). So I'm getting a lot to work with!

Whew.

Obviously, I've come to a bit of a fork in the road. My options are:

1. Move on with the 813sX813s build, and sell off the 4-1000s, sockets, filament transformer, etc. Probably later regretting the decision because the 4-1000s are absolute beauties.

2. Use a single 4-1000 as my PA and find something to modulate it, perhaps a pair of 833s. That way, I could still use the 2 10VCT 10A transformers I have on the way.

3. Something else

Option #2 seems feasible. Looking at the 4-1000 data sheet, it appears that one would be fairly happy loafing around at 2500V. Same with the 833s. I think the only difficulty might be the addition of a bias supply for the 833s, since it seems they prefer to have a bit o' bias. Not a really big deal though. This would leave me a spare 4-1000 for replacement down the road.

Now, I've never worked with 4-1000s before, so what do the gurus here think of the possibility?

[K1JJ]

Yes, the 4X1 option #2 is the way to go. It's just like buying a gun safe... go with something bigger than what you need, cuz you will find a way to fill it up.

There's many plate modulated  813 rigs on the air. There precious few 4-1000A plate modulated rigs. Half the fun of operating is telling your buds about the rig. There is a certain awe when someone tells you they are running a 4X1 plate modulated by 833A's or 4X1's.

It will take about the same efort to do 813's as 4X1's. You have the iron - definately build the big rig. Later, you can always go back and build an 813 rig when you feel green. It will feel like a pushover in comparison...

T

[WA1GFZ]

Why not modulate the 4-1000A with a pair of 813s triode connected? Or 4 push pull parallel.  Build the rig modular style so it can be upgraded when you find bigger parts. A 4-1000a at 2500 volts should be easy to tame. 

[WBear2GCR]

 4 x 813 push pull parallel might fly... you have to be careful about the balance between each pair... and you need mod iron with a low enough primary Z, I would think... since the plates will be paralleled...

Otoh a pair of 4-400 or even 4-250 will likely modulate the pants off it...

       

                   _-_-bear

[K1JJ]

Yep, I run my 4X1 modulated by a pair (all tetrode-connected) at 2200 volts most of the time. It will put out plenty of soup. It has a 3KV tap which is more than enuff if you match the tank to the final properly and draw some serious current.  I just love the 4X1 for AM work.

813's in p-p par will work OK fine, but personally, I would stick wid 833A's or a pair of 4X1 as modulators. If you're gonna build a big rig, do it right. It's the difference between riding a real Harley or a  Honda Harley. Both will do the job, but one has more class..

T

[Steve - WB3HUZ]

Bill, KB1DX has/had 4 x 813s as modulators. He sounded great!

[K1JJ]

Bill, KB1DX has/had 4 x 813s as modulators. He sounded great!

Yep, he sounds excellent. I'm the one who suggested to him to add two more 813's. He also acquired my old Peter Dahl mod iron for that rig years ago. I believe he is running 4-400A's in the final...

T

[N8UH]

I'll have to figure out the ratio on the transformer when I get it all home. The 4-1000s were modulating a 3cx2500a3. Is there a way I can figure out the ratio?

I was thinking of doing 4X813s, but figuring around $50/tube for 813s or $79 (on sale, RF Parts) for 833s, well...

That, and I like the looks of them

[K1JJ]

Yep. To figure out the modulation transformer ratio, plug the primary into 120VAC and then measure the secondary winding voltage. For example, if you get 120V in and 84 volts out, then this is a ~ 1.4 : 1 turns ratio step down. What really means something for calculating tube matching is the impedance ratio. Just square the turns ratio to get the impedance ratio. 1.4 * 1.4 = 2:1 step down imp ratio.

If you measure 120V in and 120V out, then this is both a 1:1 turns AND impedance ratio.

T

[K5UJ]

If you are planning from scratch I would go with a dual tube modulator rather than 4 x 813 simply because two tubes means less tubes to deal with if there are problems, and less hardware to buy, half the no. holes to punch and so on.

In any event I recommend going for a lot of audio power.   I would take the carrier the rig can run, and figure out the audio needed to asymmetrically modulate it to 150% (around 6 times carrier) and at least shoot for that.  You may not run that much power/audio but you will be able to go to say 100% cleanly with a lot of compression and other processing which you may get into some day so eventually you'll have the reserve built in to handle an outboard broadcast processor if you ever get one.   This will let you get beyond the typical ham and some low powered mil. gear audio limits.  The few hams I work with a healthy amount of audio punch are running rigs with parts pulled from broadcast rigs.

if the iron is from Dennis in AZ I will be interested in finding out the mod. tran. impedance as the reactor that goes with it is 30 H 1.5 A if I am not mistaken and that leads me to believe the Z is relatively low, around 2 or 3 K ohms.

rob

[The Slab Bacon]

the Iron is the hard part to find, and now you have it!! 4X1s are nice friendly tubes and pretty easy to work with. Also good used pulls are cheap and readily findable.
A single 4X1 mod by a pair of 833s would be the "Holy grail". Or a pair of 4-400 modders wouldnt be bad either. It would also look pretty sexy behind a big peek-a-boo window!!

Steve has a real nice skiz for a push-pull 4-400 AB1 modulator over on the am window website, might be worth your checking out. AB1 4-400s require a lot less audio drive than the 833s!!

Tom Vu is right on the money with ringing out the tranny with 120 to determine its turns ratio. That will also help you determine your choice of mod tubes.

                                                           The Slab Bacon

[W7TFO]

Tis I, indeed responsible. QRO!;)

[KE6DF]

Why stop with a small tube like a 4-1000?

Why not see if you can find one of these: 

[N8UH]

Thanks guys, great comments. And I agree - Build it once and build it BIG! I really appreciate things that are overbuilt and under-stressed.

Now is where I get to show my relative ignorance in hollow state circuit design 

When I get the stuff all home and rest a bit after struggling it out of the truck, I'll test the transformer and figure out the impedance ratio. That's pretty straight forward. Now, after I find that out, what next? How does one choose a pair of modulators based upon this information?

I'm really interested in learning how a transmitter like this is designed. If I can better understand how and why these design decisions are made, I'll take comfort down the road knowing that the decisions made were the right ones. And it makes for good, lively discussion! 

[The Slab Bacon]

Tim,
       A good "rule of thumb" is to have modder tubes that total up to the same dissapation as the final (or something close to it). Also, you should have as much audio power as you expect to put out in carrier power. The rule of thumb is to have as much audio power as you have DC input to the final. But having more gives a nice headroom cushion so you dont have to push the audio section into distortion.

Example: 2x 833=900w dissapation + 1x 4-1000A = 1000w dissapation = close enough to be love at first sight and live happily ever after.

[K1JJ]

10-4 there Slab Bacon!

Another technique the Tron taught me for matching tubes to the iron....

If you modulate ONE final by a pair of the SAME tube type, (ONE 4X1 modulated by a PAIR) the transfomer impedance ratio required is about 1:1.   If you want to create a 2:1 step down ratio to best match a particular transformer, then use a PAIR modulated by a PAIR of the same type.  (two 4X1's modulated by a pair)

You will need naturally be running each of the two finals at their normal currents, thus generating 1/2 the impedance of just one tube.  (and putting out X2 power) However, I have loaded a pair lightly and even though the ratio may be off, there is still plenty of headroom when using 2X2 combo of the SAME type tubes throughout. It may be overkill  when using 1X2 of the same tube, but what the heck, it's super clean with lots of headroom.

I happen to have a ~1.5:1 RCA BC mod transformer, so used  one 4X1  modulated by a pair. 

The lower ratio will usually give higher audio peaks at the expense of more plate color on the modulators, assuming the tube headroom is available.

The same rule works for 813's, 833A's etc.


Of course you can use different tubes and look up the p-p tube's modulater impedance in the handbook - and calculate the E/I load of the class C final. However, the class C final is not exactly E/I - I'd have to look up the factor - or maybe someone here knows it. (0.70?)

T

[The Slab Bacon]

the turns ratio is actually more important than the impedance ratio. The impedance seen by the modder plates is a direct function of the turns ratio and the operating
parameters of the final as the munky swings.

Tom,
      we have always discussed the 1:1, 2:1 thing. 1:1 was always said to be for "1 of something modded by 2 of something similar". 2:1 was for "2 of something being modded by 2 of something similar". Just like you said. 

A little extra leverage gives more munky swing without working things as hard.

[WA1GFZ]

or when you need to move something heavy.

[Opcom]

Let's look at pp-par 813's vs pp 833's from the transformer's view.
813's are ICAS, but CCS gives 2250VDC, 10K p-p, Z and 760W
833's at 2500V in the same socket ought to do 1100+ watts

Type	Eb	Ib(max mA)	Po(W)	Zo(p-p)	Drive (W)
813	2500	580	980	9500	0
833	3000	750	1650	9500	20

If you will only ever run 2500V, 813's ought to do OK. What pleases you for appearance and reliability is what matters.

[KM1H]

I dont understand why some have to use CB sounding language....it sounds retarded, not cute, as many have commented on other forums while Googling for AM info.

Carl

[The Slab Bacon]

I dont understand why some have to use CB sounding language....it sounds retarded, not cute, as many have commented on other forums while Googling for AM info.

Carl

Because we can    

Why do we use ham jargan instead of talking like we do in non-ham circles??

Because we can!    

Life is by far too short to waste precious time worrying about stuff like that.
Life will also deal you lots of occasions to get you really pissed off, so why waste them
worrying about little stuff!

[KC4VWU]

Yeah, Frank's in good spirits; his bud is getting in good shape and he's happy, just having a little fun. Cut him a little slack.

[N8UH]

Yup, no harm in a little silly-talk. Keeps things from getting too dry

The examples and explanations above have really helped me get my head around this. I guess where I have flexibility is the fact that I'm getting 2 4X1s, so depending on mod ratio, I can use 1 or 2 in the PA to make it work. To get things exact, it's down to selection of modulation tubes for dissipation and P-P impedance. *crosses fingers hoping that 833s will work*

Since the original TX was using 2-4X1s modulating a single 3cx2500, could one assume that the ratio would be closer to 1:1?

Slab, I see what you mean by having enough audio - matching modulator dissipation to PA dissipation seems like a good rule-of-thumb to give you more than enough audio.

As I mentioned before, I'm getting the original audio amp section from the TX - 12BY7s driving EL34s that were slapping the 4X1s around cathode-follower style. I'm thinking that with a bit of work it should work fine as an audio driver for what ever I have in the modulator section. But I'm going to have to change out the EL34s to something the AudioPhools haven't driven the price up on. So I'm probably covered for audio drive. I'm not sure if I'll need to switch over to transformer coupling though.

Now, considering the use of a 4X1 in the PA, how does one figure out what will I need for exciter power?

[The Slab Bacon]

"Now, considering the use of a 4X1 in the PA, how does one figure out what will I need for exciter power?"


The rear appendix in an early ARRL handbook is a good place to look for operating parameters of older transmitting tubes. I keep a 1963 handbook on the workbench just for reference. IIRC a single 4X1 needs 25-30w RF input to the grid to get happy.
A little extra grid drive will also never hurt when trying to make the high audio peaks.
I drive mine with a single 6146 with somewhere around 450-500v on the 6146 plate.
(around 45-50w DC input) or somewhere around 23-30w of drive.

[K1JJ]

Tim,

Slab is right on with the drive requirements in class C.  Depending on whether you neutralize the final or swamp the grid with a non-inductive resistor (like 2K-5K) will dictate the actual RF drive needed. The bottom line is getting the 40 ma grid and 140 ma screen current under full operating plate load. I drive mine with maybe 15-20w of dead carrier from the ricebox.  Mine is neutraized and has about 8K of swamping, so easy to drive.

For an audio driver, consider the all-new WA1GFZ MOSFET driver. I use one to drive the p-p 4X1 modulators in tetrode connected class AB1.  It's as clean a driver as you will ever get with response so low that you can just about run shark fin audio polarity. The only limitation is the mod iron.  A direct coupled tube driver (I think what you were discussing) will be FB too, of course.

Info and pictures of the GFZ MOSFET audio driver prototype for most any modulators, including 811A, 813's, 833A's 4-1000A's, etc are below.

http://amfone.net/Amforum/index.php?topic=23632.0

T