KA9EGW
Member
 Offline
Posts: 79
|
 |
« Reply #25 on: September 23, 2013, 07:54:42 PM » |
|
Dropping Ep to 2200 is more real world and makes the load Z more like 3.5K...I have enough core E and I laminations around here I could probably wind my own modulation transformer...I figure [hope] it doesn't have to be any [much] bigger than the plate transformer, core-size-wise...one spare core is 5.5" wide, 3.5" thick and 6.5" tall...if the windings have to be universal-wound I'll have to fabricate a cam for the lathe, but that is doable...that is if the bobbin won't fit in my Morris coil winder...I figure Bakelite would be a good material for the bobbin [that core is currently potted so fire is relevant]...
|
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #26 on: September 23, 2013, 09:21:06 PM » |
|
I would stick to 2000 volts and 300 to 350 ma.
There really is no point in pushing things so hard.
The difference on the receive end is going to be very slight.
|
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #27 on: September 23, 2013, 09:40:15 PM » |
|
300-350 for the pair, yes? And you're right, realistically this will be good for at best 1 s-unit over my 100-watter. I'm in it at least as much for the fact I just plain like building stuff as anything...  |
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #28 on: September 23, 2013, 11:05:38 PM » |
|
A pair of 813's in a plate modulated class C will do 600 watts of carrier output at 2000 volts at 400 ma.
About 450 watts at 2000 volts and 300 ma.
Both are over the legal limit.
When you go above 2000 volts, the modulated voltage spikes can be very high.
6000 or 8000 volts in some cases.
Things tend to get large at higher voltages...
|
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #29 on: September 23, 2013, 11:19:27 PM » |
|
Then they should have no problem whatsoever doing 375W...which @100%, works out to 1500W PEP. I could run the two of them at ~1.4kV and still get my 375W out...easily. Tubes ought to last about forever that way. I SO miss the days when 1kW DC input was the limit...  |
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #30 on: September 24, 2013, 03:33:32 PM » |
|
I am not sure many people pay attention to power limits... 2000 volts at 300 ma would have the tubes last forever. I have had the same two 813's in the big rig for the last 25 years. I just picked two out of the pile and have been using them since. I never really tried what the maximum power out is, 700 or 800 watts carrier I suppose. I sent a box out today with a bunch of extra stuff in it, along with the choke. Supposed to be there Monday. Use it, give it away, put it in the trash, its up to you. Any idea what you are going to do for modulators? 811a's are kind of wimpy and low voltage for a paair of 813's. Then they should have no problem whatsoever doing 375W...which @100%, works out to 1500W PEP. I could run the two of them at ~1.4kV and still get my 375W out...easily. Tubes ought to last about forever that way. I SO miss the days when 1kW DC input was the limit... |
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #31 on: September 24, 2013, 03:44:19 PM » |
|
Modulator? Heck, I'm still working on the power supply I agree 811's would be a bit underpowered...I'll have to do some research; maybe another pair of 813's? After all the 813 sockets DID come in a 4-pack...maybe a single 833A? Dunno yet. |
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #32 on: September 24, 2013, 05:42:47 PM » |
|
How would you use a single 833 as a modulator?
I never did any experiments with 813's as modulators.
You can triode connect them or not.
Good power out in AB2.
|
|
|
|
|
Logged
|
|
|
|
K1JJ
Contributing
Member
Offline
Posts: 8893
"Let's go kayaking, Tommy!" - Yaz
|
|
« Reply #33 on: September 24, 2013, 06:24:13 PM » |
|
I've used 813 triode-connected modulators, driven by a pristine MOSFET audio driver. The Tron once told me it was the best sounding rig I ever had. The THD and triangle tests looked very good as well. Though, I'll bet that a pair run AB1 as tetrodes with regulated screen and grid supplies would be even cleaner. I've done both configs with 4X1s but never tried 813 tetrode-connected. The most I've even loaded up a pair of 813s in class C plated modulated service was 1KW out using 3KV. The 813 modulators were able to do 130% modulation at that power level. All in all, a pair of 813s X a pair of 813s is a very nice lash up. They require little air, the filaments are robust and the tubes are common and cheap. A 2:1 step down mod transformer when using a common supply works nicely. A quad set X a quad set wud be outstanding. Almost built a quad rig once, but opted for the wimpy 4X1.....  T |
|
|
|
|
Logged
|
Use an "AM Courtesy Filter" to limit transmit audio bandwidth +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed. Easily done in DSP.
Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."
There's nothing like an old dog.
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #34 on: September 25, 2013, 01:36:20 AM » |
|
Brett, I'm really not 100% sure how I'd use an 833 as a modulator; at this point I merely note the tube is about a hundred bucks and attractive from a $-per-watt standpoint HI
73, Brian KA9EGW
|
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #35 on: September 25, 2013, 08:03:22 AM » |
|
Most modulators use 2 tubes in push pull.
Two 833's take a LOT of power for the filiments I think.
|
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #36 on: September 25, 2013, 10:35:29 AM » |
|
ONE 833 is tempting but it's have to run single-ended Class A. Not real efficient. Not even covered in any datasheet I can find. TWO 833's [plus the necessary air conditioner for my heat-intolerant XYL] would cause the bearings in my electric meter to burn out [and land me in divorce court] I suppose 813's modulated by 813's is a well-proven technology...like I said I'm still figuring out that part of the design, tossing all the ideas at the wall to see what sticks... |
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #37 on: September 25, 2013, 11:43:42 AM » |
|
Class A would be very cool.
I never tried that, can you just use a tube as a pass tube, running 2x the plate voltage and dropping half in the pass tube, then modulate the pass tube grid?
No mod transformer....
|
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #38 on: September 25, 2013, 12:40:53 PM » |
|
The first place my mind went when I read this, was back to my 1947 Handbook...submitted for the group's consideration that a pulse-width-modulated pass tube setup and a vacuum-tube-keyer setup are, if not littermates, at the very least of the same gene pool...and the term "cathode modulation" came to mind...the whole mental image is still fuzzy at this point; a very quick scan of the '47 ARRL HB just shows a mod xfrmr secondary in the final amp cathode ckt...more research is indicated...I'm awaiting my copy of Orr's HB from the mail order bookseller...
|
|
|
|
|
Logged
|
|
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #40 on: September 25, 2013, 01:43:17 PM » |
|
Thank you!
|
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #41 on: October 09, 2013, 05:23:13 PM » |
|
You guys think 120uF/3kV is enough?  |
|
|
|
|
Logged
|
|
|
|
KA2DZT
Member
Offline
Posts: 2192
|
|
« Reply #42 on: October 10, 2013, 02:10:03 AM » |
|
You guys think 120uF/3kV is enough? Probably three times what you need. 40ufd is more than enough. Fred |
|
|
|
|
Logged
|
|
|
|
N2DTS
Contributing
Member
Offline
Posts: 2307
|
|
« Reply #43 on: October 10, 2013, 09:29:31 AM » |
|
You have to be careful how you apply power to that much UF. In my modulator supply, I have choke input, 30UF on the chassis, and a 40 uf on the floor of the cabinet. Four 1 amp 17kv rectifiers (2 amps per leg) and step start. The RF deck just runs a wimpy 30UF. The chokes are UTC cg series rated at 500ma 7000 volts ccs, the mod deck is a swinger. You only need a lot of UF for a modulator supply. If you use a common supply, build it for the modulator I suppose, lots of UF. My 3x4D32 and 2x4x150 rig runs a common supply, because I an to lazy to build another, and it runs 0-1400 volts, 80UF of caps, choke input step start. The power transformer is a Collins solder sealed job good for 560ma I got from Fair Radio back when they had lots of good parts cheap. You guys think 120uF/3kV is enough? |
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #44 on: October 10, 2013, 10:36:02 AM » |
|
Oh yes, that much uF would look like a dead short for a significant period of time. I'm definitely *not* ruling out some sort of current-limiting scheme. The very simplest is an LM317 with the output tied to the center leg through an appropriate power resistor and the whole affair left floating, although I fear that although the concept is somewhere between valid and necessary, that execution is not appropriate at the voltage involved... As always in my world, more research is indicated... |
|
|
|
|
Logged
|
|
|
|
KA9EGW
Member
Offline
Posts: 79
|
|
« Reply #45 on: October 10, 2013, 11:33:06 AM » |
|
NTC thermistors. That sounds viable...
|
|
|
|
|
Logged
|
|
|
|
|
