Modified Heising blocking capacitor question

[k4kyv]

With that gap, it should be designed to  carry the DC through the transformer secondary.

I think mine, with no gap, is marked XT-1620.  I'll have to check for sure next time I am out in the shack.  One of mine came from a rig that used a pair of 805's modulating another pair of 805's.  George, W1UAX used to use one of those transformers in his rig.  No gap, and I don't think he used a reactor, but his rig didn't have a lot of low end.

I have another mod xfmr designed for an RCA 250w transmitter without a reactor.  It has a gap, but is much larger; about the same size as the early 1 kw RCA transformers that used a reactor.  In is rated about 15000Ω:5000Ω.  Not a good ratio if you are looking for positive peak headroom, unless you use separate power supplies.  With that high a primary Z, it may be designed for the 828 modulators.

Ask Timtron.  He looked it over one time when he was down here for a visit, and was familiar with the iron and the transmitters they went in.

[w1vtp]

I can read it on the photo.  I forgot about clicking the little + sign to enlarge the image.

Interestingly, 10,400Ω is the rated p-p load impedance for a pair of 833A's or 849's.  I'm sure that transformer is not rated for the power level of a pair of those tubes. The one I saw at Dayton had a manual from RCA that included all the characteristic data on the transformer, even down to a diagram of how it was wound, the size wire and thickness of the insulation.  I didn't need the transformer so I passed it up.  Hope someone who could use it bought it.  It would have been a shame for it to go to the dump.

I would REALLY be interested in any data I could get for the RCA beast.  I'd love to use it in my legal limit xmtr.  I do have a CVM-4 I can use if I have to.

Any ideas on Mod tubes vs RF tube that would fit this mod iron?

Al

[Opcom]

Looking at the modulator side, which is the hardest side to match up, 10.4K CT with a HV of 1700V is what I am having issues with.. If the iron is rated 2500V test.. It was likely made for the hallowed old tubes of the day, now rare. Here are some possibilities from the old books.

Type	B+	p-p load	output	Secondary Z	comment
	(VDC)	(Ohm CT)	(W)	(Ohms)
RK-18	1250	9000	380	3764	4 tubes pp par. Raytheon Tube
RK-31	1250	9000	380	3764	4 tubes pp par. Raytheon Tube
35TG	2000	13750	470	5751	4 tubes pp par.
TZ-40	1500	12000	250	5019	Taylor Tube
HY-51B	1000	9000	285	3764	Hytron Tube, 10V/2.25A (HY-51A is 7.5V/3.5A)
RK-52	1250	10000	250	4182	Raytheon Tube
4-65A	1800	10000	540	4182	4 tubes pp par.
75TH	2000	9650	600	4036	4 tubes pp par.
GL-146	1250	8400	250	3513	G.E. Tube
GL-152	1250	8400	250	3513	G.E. Tube
HD-203A	1750	10000	400	4182
203-A	1250	9000	260	3764	G.E. or RCA Tube
203-Z	1250	8000	300	3346	Taylor Tube
211	1250	9000	260	3764
HF-300	2000	9600	650	4015	Amperex Tube
311	1250	9000	260	3764
HK-354E	2000	11000	472	4600	H&K Tube
T-805	1750	9350	320	3910	Taylor Tube
805	1500	8200	370	3429
811	1500	9000	450	3764	4 tubes pp par.
811A	1250	9300	310	3889	ICAS
811A	1500	12400	340	5186	ICAS
813	1500	9300	260	3889	Class AB1 CCS
813	2000	8000	670	3346	4 tubes, pp par., Class AB1 CCS
813	2250	10000	380	4182	4 tubes, pp par., Class AB1 CCS
813	2500	9500	980	3973	4 tubes, pp par., Class AB1 ICAS
828	2000	9250	770	3868	4 tubes pp par.
GL-835	1250	9000	260	3764	G.E. Tube
838	1250	9000	260	3764
5514	1500	10500	400	4391
7094	2000	12000	560	5019
8000	2250	12000	725	5019
8005	1500	10000	300	4182

If there is too much power, just don't drive it past what is needed.

That said, where a quad of tubes can make 770W and one only needs 450W, there is no reason that one pair cannot be used into the same transformer impedance and just push the single pair a little harder, into ICAS region. The determining factor is whether they can deliver the current. Wierder things have been done.

For example in the audio charts (by the book) the 811A at 1250V does 235W into 12500 Ohms CCS and 310W into 9200 Ohms ICAS.
The book also calls for 1000V and 5100 Ohms CT for 178 watts CCS. The Altec 1570B audio amp runs 920V and 6500 Ohms to make 175W which is the opposite of what one would expect.
I mean by this that this stuff is pretty flexible up to the point the wrong anodes turn red.

Maybe someone has a case of several of these old tubes around with no use for it. That's how I found my spare 673 rectifiers (thanks to Barrie who had some in very good condition.)

One plate to plate load I did not find is a pair of 813's in the zero bias class B triode circuit, where the screens are driven. That might be a good possibility.

I put this in excel and saved as a comma delimited text file so it can be sorted in excel how you like, if you think it's worthwhile. (note to mods: please consider allowing excel files, or zipped excel files..)

[Opcom]

there is more to it as well, to get the most out of it without cooking it.. The PA conditions have to be taken into acct. as to what is safe for the xfmr, or at least what is within its original ratings, it is rather old. From those data, there will be choices for the RF amp tube. so I am cooking that up. Maybe for my own amusement as much as anything.. addicted to working AM problems in excel at 4:39 AM.

[k4kyv]

The p-p load of the modulator tubes is not critical.  Those listed values are what is considered optimum, but as long as the maximum voltage or peak current is not exceeded I wouldn't worry about using a different load impedance, +/- 30% or so.

The nominal load impedance on a transformer is even less critical than that of tubes.  You should be able to go +/- 100% (from one half to double the nominal impedances at the same ratio) without any noticeable degradation in performance, again keeping maximum rated voltages and currents in mind.  The important thing is turns ratio, which defines impedance ratio.

If you are using a common power supply for modulator and final, the turns ratio needs to be somewhere in the ballpark of 1.4:1, or 2:1 impedance step-down.  Too much step-down will limit the positive peak capability, and too little step-down will cause the tubes to run inefficiently, and if you try to force feed them in too low a p-p load impedance, excessive distortion may occur.

Very often, the fixed turns ratio on military and commercial transformers will be close to 1.7:1 (3:1 impedance ratio).  This ratio limits modulation percentage to about 100% if a common power supply is used, making it impossible to substantially overmodulate the transmitter, although the same kind of distortion and splatter will still occur due to the clipping action of the modulator tubes.  This turns ratio limits the positive peak capability of the BC-610, the older 1 KW RCA broadcast transmitters, and the the 250-watt RCA rig that used the huge mod xfmr with no reactor (that particular transformer is nominally 15K:5K).

"High level" speech clipping operates on this principle, with a  low-pass splatter filter between modulation transformer and load. Unfortunately, a fatal error in using high level clipping with a splatter filter is that it tends to blow modulation transformers when a strong audio tone occurs right at the cutoff frequency of the filter.
 
A good example of using a given turns ratio to cover a wide range of impedances is the case of the so-called "multi-match" modulation transformers like the UTC VM or CVM series.  If you look at the charts, you will notice a limited number of turns ratios available using various tap configurations, and the sequence of turns ratios is repeated over and over again throughout the impedance range.  For instance, a given tap configuration may give a ratio of 10K to 4K.  If you examine the chart you will notice the exact same tap configurations for 20K to 8K, 12K to 4.8K, 8K to 3.2K, etc.

Whenever I have used multi-match transformers, I always used the entire primary and secondary windings and operated the transformer as a fixed ratio transformer.  Typically these run somewhere about 1.3:1 turns ratio when the entire windings are used with all sections connected in series.  I used the winding with  the higher number of turns as the primary.  The only thing I used the chart for was to make sure I had the winding sections connected with the proper polarity, since reversing the taps that are supposed to go to the ends of the windings with the taps that are supposed to be strapped together to form a mid-tap may blow the transformer due to insulation break-down.

[w1vtp]

Hi Opcom

I did download ur CSV and converted it over to Excel.  Thanks.  I vaguely remember 4-400's being associated with this xfmr.  Does that make any sense?  Anyway, more than half the fun is figgerin out this stuff

Don

I'm going to save your comments in MS Word and reread it several times.

Thanks, Al

[N2DTS]

Yes, I have 2 of those transformers, got them new in the crates.
MUCH better than the CVM-5!
They seem to sound fine with DC, I use one in the 3x4d32 rig and one in the 2x813 rig.

Both rigs run around 5000 ohms on the RF side, and something close on the mod side, which worked out well...
Since I can variac the voltages seperatly, I can change the impedances but dont hear any big difference.

When I tried the modified hy-zing, I would not say things sounded better, different but not better.
A wide range of caps were tried, in both setups, iron hot, iron not, they all sounded different, but I dont think anything sounded as good as without....

Best mod iron I ever used, seems good for max power, about 700 watts carrier, 600 watts audio power.
When my meter shunt went south, I was likely running 2500 volts or more on the 813's at 400 ma, with plenty of audio, and the trans did fine.....
What clued me in was the smell, like hot metal and glass, plus getting 700 watts out with 800 watts in showing....


Brett

When I tried this setup, the capacitance had a large effect on the audio sound, every amount of capacitance sounded different.
I gave up and just use the RCA mod iron with current through it...


Brett

Interesting comment Brett.  I think I have the same mod iron that you are using

Al

[W2XR]

I have what I believe to be the same RCA modulation transformer.

The RCA dwg. number on the nameplate of this unit is 900777-502. The impedances are 15,000 ohms plate-to-plate on the primary side, and 5030 ohms on the secondary. It came out of a 250 watt RCA BC rig using a pair of 828s modulating a pair of 810s. The B+ was 1600 VDC. I recall the transmitter model number as being an RCA Model 250K.

This transformer weighs in at about 150 lbs, and is of open-frame construction.

I used this transformer in my first homebrew high-power plate modulated AM rig, which I built back in the late '70s. It used a single 4-400A with 2750 VDC on the plate at 750 watts DC input, modulated by a pair of class B 833As, and no modulation reactor; it didn't need one, as the mod tranny was gapped for the unbalanced DC. It just barely made 100% positive peaks due the gross impedance mismatch, but the transformer had incredible audio bandwidth and fidelity. I don't recall the swept audio response, but the low-end in that thing was liquid, as TimTron used to say.

I actually had two of these RCA modulation transformers at one time; both came from the same source. I kept one, and gave the other to Jerry, WA2FNQ. He used his also with a single 4-400A, but with a pair of 572Bs as modulators.

73,

Bruce

[N2DTS]

Oh, and as far as tubes you can run with that iron, there are loads of choices.

Pair of 813's runs 5000 ohms at 2000 volts, 400ma.
Three 4D32 tubes runs a little less at 1500 volts and 310 ma.
Push pull 812a's would be 1500 volts at 300 ma, 5000 ohms.
Lots of tubes like the 4-125/400 could do 2000 volts at 400 ma.


For modulators, 811A's at 1500 volts zero bias work well, as do 4cx250b (2000 volts, 500ma, 600 watts), 4 100th tubes are close at 2000 volts, etc.

Almost any tubes will work if you pop one more in, or reduce the voltage, etc.

I did run the same transformer in the push pull 812A rig modulated by 811A's, all matching RCA iron, and that rig also sounded very good.

Brett


It seems to work very well over a wide range of impiedances.

  

[W2XR]

  This turns ratio limits the positive peak capability of the BC-610, the older 1 KW RCA broadcast transmitters, and the the 250-watt RCA rig that used the huge mod xfmr with no reactor (that particular transformer is nominally 15K:5K).
/quote]

I have what I believe to be the same RCA modulation transformer.

The RCA dwg. number on the nameplate of this unit is 900777-502. The impendances are 15,000 ohms plate-to-plate on the primary side, and 5030 ohms on the secondary. It came out of a 250 watt RCA BC rig using a pair of 828s modulating a pair of 810s. The B+ was 1600 VDC. I recall the transmitter model number as being an RCA Model 250K.

This transformer weighs in at about 150 lbs, and is of open-frame construction.

I used this transformer in my first homebrew high-power plate modulated AM rig, which I built back in the late '70s. It used a single 4-400A with 2750 VDC on the plate at 750 watts DC input, modulated by a pair of class B 833As, and no modulation reactor; it didn't need one, as the mod tranny was gapped for the unbalanced DC. It just barely made 100% positive peaks due the gross impedance mismatch, but the transformer had incredible audio bandwidth and fidelity. I don't recall the swept audio response, but the low-end in that thing was liquid, as TimTron used to say.

73,

Bruce

[Opcom]

Hi Opcom

I did download ur CSV and converted it over to Excel.  Thanks.  I vaguely remember 4-400's being associated with this xfmr.  Does that make any sense?  Anyway, more than half the fun is figgerin out this stuff

Don

I'm going to save your comments in MS Word and reread it several times.

Thanks, Al

I had alot of fun doing that. Sometimes I dig too far into things. I started to expand the spreadsheet and found that some of the situations could result in up to 150% (voltage) modulation, which is 225% modulation looking at power, and some questionably high winding voltages. But the thing would be to not drive the mod that hard.

That brings up a point. When we here speak of 150% positive modulation:

Do we mean that a 50V (50 watt) carrier as shown on the scope now is at 150V peaks (450W) instead of the 100V peaks (200W) we would see at 100%?

Or,

Do we mean that the peak power is 150% of the "100% mod" peak power? (-that is the 50W carrier's PEP is 200W at 100% mod, and 300W at 150% positive modulation.) That is quite a difference.


I think the 4-400's were in a more modern 1KW transmitter, whereas the labeled iron I showed, was from the 1947 style 250W rig with 828's modulating 810's at about 1700V. (looks like what Don said). Overvoltage would probably be more prone hurt it more than overcurrent. The matching KW stage for the '47 transmitter (BTA250K or L) used 833's and IIRC 2500VDC. All the transformrs in the pictures, except the large "unidentified transformer" I posted, seem to be the same size, like a transformer from a 250W BC rig. That said, BC fidelity at 250W could be equated to speech fidelity at 1KW, if the transformer does not burn. Just my opinion on it.

[k4kyv]

That brings up a point. When we here speak of 150% positive modulation:

Do we mean that a 50V (50 watt) carrier as shown on the scope now is at 150V peaks (450W) instead of the 100V peaks (200W) we would see at 100%?

Or,

Do we mean that the peak power is 150% of the "100% mod" peak power? (-that is the 50W carrier's PEP is 200W at 100% mod, and 300W at 150% positive modulation.) That is quite a difference.

By 150% positive modulation I mean the modulation percentage is 150%.  If there is 1000 volts DC on the RF final, at 100% modulation you are adding another 1000 peak a.c. volts, raising the total peak voltage to 2000 volts.  At 150% positive peak modulation, you are adding 1500 peak a.c. volts to the 1000 volts, raising the total peak voltage to 2500.  At that point, the peak power on positive modulation peaks would be 6.25 times the unmodulated carrier power.  If the unmodulated carrier power is 50W, the instantaneous peak power would be 312.5 watts.