Homebrew 4-1000A plate modulated by a pair - Questions

[AB2EZ]

Tom

The output power (at 100% sine wave modulation) increases to 4000 watts on positive modulation peaks... but it also drops to 0 watts on negative modulation peaks.

When you average over one cycle of the modulation [and you need to do the averaging carefully, using the fact the the power is proportional to the square of the voltage waveform... and the voltage is 1+ cos (2pi ft), where f is the modulation audio frequency]...

You find that the average power is 1500 watts. 1000 watts of power coming from the B+ supply and 500 watts coming from the modulator.

Also note: since the rf stage is probably operating at around 75% efficiency, and the modulator has to deliver 50% of the input carrier power, the required audio power is 500 watts / 0.75 = 667 watts of audio

Looking forward to hearing more.

Stu

[K1JJ]

Thanks for the replies.


OK, I understand the AVERAGE power required from the modulator. That's what hams have been using for years and is probably the reason why many plate modulated rigs flat topped and ran out of headroom.  I think PEP power is a better indication of the modulator requirements.

Let's talk in terms of pep power that the modulator must suppy at any one instant. This is where the damaging voltages which crap out transformers occur. The average power parameter is OK for measuring current and heating effects. But peak voltages and core saturation are greatly affected by peak power.

So if we have a 1000w constant AM carrier input, how much PEAK power is required from the MODULATOR to modulate 100%?    

I'm trying to deteremine just how much peak power I can pull out of this mod transforer before worrying about flashing it over. It's big enuff to handle 1000w of average power, in my opinion, but at 3500V using 4X1's the peak voltages could get out of hand depending on the peak power required, or mistakes made.  It has a "1 KVA" tag on it.


Brett: Good on the RCA transformers. I'll post some pics of the one I have and see if you are using the same ones. Yes, tone tests can be really stressful. I've blown only one mod transformer here over the years. It was a 500W BC unit when I was sweeping it at about 1KC. Suddenly the secondary opened up and that was it. Really sucked.  So I try to limit the % of modulation  when testing, but that is hard cuz we need to know how the extreme peaks act at 120%, etc.

[W2VW]

I just moved the brother to that REAL mod xfmr you had Tom. It lives in the crawl space below one end of the house now. I used an engine crane to lift it high enough to fit into the entrance.

Why not suspend your mod iron in oil and hook the thing up with the series cap on the cold end of the secondary?

[K1JJ]

Another thing I just found out....

Derb was nice eniff to send me an old 1951 manual regarding big tubes.  One chapter was on cooling. In the past I advocated drilling holes in a circle around the inside chimney. That's OK for 813, etc - tubes w/o chimneys, etc. But for chimney tubes it says not to do this OR do not sub-mount the tube. This applies to tubes with chimneys like the 4-400, 4X1, 3-500Z, etc.  The air is supposed to pass only thru the seals base holes on its way to the plate glass.

I remounted my 4X1's flush and resealed the holes. The air flow is somewhat reduced, as expected, but it is now flowing through more critical areas, like the seals.  The added back pressure also equalized out the flow between tubes, so a plus there.

Thanks for the dope, Derb!

BTW, Stu, one of the 4X1 modulator tubes I'm using is one I got from you. It's one of the best of the bunch and is a good match for its mate.

T

[N2DTS]

PEP audio power would be half the carrier power, plus whatever asymetrical positive peaks you have (say 10%), plus losses in the mod transformer (10%?).
My guess would be 600 or 650 watts of audio.
Average power will be much lower, unless you transmit sine waves....

Brett


 

[WA1GFZ]

Tom,
Now that you have the tubes mounted correctly measure the air pressure under the tubes to see if you have enough flow through the tube bases.
Glass tubes like the 4-400 and 4-1000A need a fair amount of back pressure to cool them properly in AM service. My old 4-1000A rig I pressurized the chassis and pulled a vacuum on the top side and was still a bit light but was ok at linear duty cycles.

[K1JJ]

PEP audio power would be half the carrier power, plus whatever asymetrical positive peaks you have (say 10%), plus losses in the mod transformer (10%?).
My guess would be 600 or 650 watts of audio.
Average power will be much lower, unless you transmit sine waves....

Brett

I'm still not clear on this...

If a peak reading wattmeter shows 1000w carrier and 4KW pep at 100% modulation, the peak power has to be coming from somewhere. If the mod xfmr is in series with the plate circuit and is adding say 3500V to the series plate circuit, and the series current is the same, then it is responsible for its 2,000 watts peak power share of the circuit, no?

IE, It seems to me that if the mod xfmr secondary is in series with the plate circuit (no Heising in this case) it is delivering the full 3500v and a full series current of the plate circuit when the power hits 4KW pep. This equates to ½ of the peak power, which is 2KW peak from the modulator.

Someone set me straight before I hang myself, please.  

T

[N2DTS]

Well, if you increase the voltage, current does not stay the same, its not regulated....
So figure when the plate voltage doubles, the current doubles as well, 4x the power.

This all happens so fast, the average does not change much, and is offset by the power going down to almost zero on neg peaks.

On my rigs, the meter movements dont change under voice modulation, rms rf output power might increase a little, pep goes way up...

Brett

[K1JJ]

Yes, but regardless of the duty cycle or power coming down to zero throughout it's cycle, what is the peak power of the modulator's output? If the modulator cannot supply this peak power it will flat top and saturate, causing distortion.  It's just like a linear amplifier.

Actually, to put out 2KW peak modulator power, the modulator tubes have to take in about 3.3kw at 60% efficiency to modulate an input power of  1333w carrier (75% final eff).

This peak power requirement means the power supply has to handle the heavy dynamic regulation as well as the mod xfmr core being able to handle and pass the instantaneous peaks.

I just need someone to admit that the peak power coming from the modulator is 1/2 of the peak power of the RF envelope. IE, 4KW peak RF output = 2KW modulator's share.  Or show me why it's not without math formulas.


BTW, Brett - I saw your post about not finding the plate meter of much value in a linear. I see the same thing. The tank tuning looks lazy, but not if you're watching the power output meter.. I use the power output peak reading meter as a guide.  You can see everything on a scope and power meter to load it correctly. I use the plate current meter mainly to set and monitor the idling current of the finals.
T

[N3DRB The Derb]

I have always visualized cooling air as just like flowing water. If you need water to go in a certain spot, you direct it there and make sure you have enough force to get enough of it across the target, but getting more flow by widening your 'drain' size does not help much because all you are doing is giving the cooling agent more of a chance to miss your target.

[AB2EZ]

Tom

Let's put out a few numbers. To keep it simple, let's assume that with only carrier (no modulation), the B+ is 3500 volts, and the plate current is 400 mA. Therefore the power is 1400 watts (input).

Now let's add 100% sine wave modulation.

And (important) let's assume (temporarily) that we not using a Heising configuration. We'll come back to the Heising configuration after we consider the non-Heising configuration

On positive modulation peaks, we have the following:

The plate voltage is 3500 volts x 2 = 7000 volts
The plate current is 400mA x 2 = 800 mA.

The total input power is 5600 watts (i.e. 4 x the input power when there is no modulation).

The B+ supply is providing 3500 volts at 800 mA = 2800 watts (2 x the input power it provides with no modulation)

The modulator is (also) providing 3500 volts at 800 mA = 2800 watts

Thus on positive modulation peaks, half of the total input power (2800 watts) is coming from the DC supply. The other half of the input power (2800 watts) is coming from the modulator.

Since the DC supply is providing 800 mA, its output capacitor is discharging twice as fast as it does when there is no modulation. But, that's not a problem provided it isn't going to discharge very much is in the time it takes for the modulation to complete 50% of one audio cycle. If it does discharge (and recharge) more than a few percent in each audio cycle... at the lowest audio frequency you care about... then you need more capacitance at the output of the B+ supply.

Now let's consider the negative peak of the modulation cycle.

The B+ is 3500 volts. The modulation is -3500 volts. So the total voltage on the plate of the rf output tube is 0. The modulator is putting out 0 mA. The power supply is putting out 0 mA. The r.f. output tube has zero plate current.

So, for the non-Heising case, the peak voltage across the secondary winding of the mod transformer is 3500 volts. The peak current (that occurs at the same time as the peak voltage) through the secondary winding of the mod transformer is 800 mA.  The peak power that is being delivered by the mod transformer secondary is 3500 volts x 800mA = 2800 watts (as stated above).  But, this number can be very deceptive and/or confusing, because half of that peak power (1400 watts) is being delivered during the positive half of the modulation cycle by the stored energy in the transformer, and only the other half of that peak power (1400 watts) is being delivered by the modulator tubes to the primary of the transformer. This will become clearer when we "discuss" the Heising case (next).  

Next, let's consider the case of Heising modulation.

On modulation peaks, the peak voltage across the secondary of the mod transformer is 3500 volts (same as non-Heising case)

The peak current flowing through the secondary of the mod transformer is 400 mA (half as much as in the non Heising case).

The peak power being delivered by the secondary of the mod transformer is therefore 1400 watts (half as much as in the non-Heising case).

The B+ supply is delivering 3500 volts and 400 mA... independent of the modulation peaks. So the B+ supply is delivering a constant 1400 watts.

So.... if the modulation transformer secondary is delivering 1400 watts on modulation peaks (both positive and negative), and the B+ supply is delivering 1400 watts (all the time), and if the total input power to the rf stage is 5600 watts on modulation peaks... then: where does the remaining 2800 watts come from (in the Heising configuration)?

Answer: it comes from the energy that is stored in the magnetic field of the Heising choke (E=1/2 LI**2), and the energy stored in the electric field of the Heising capacitor (E= 1/2 CV**2).
 
The Heising choke is always carrying 400 mA (except a low modulation frequencies). But, on positive modulation peaks, there is +7000 volts  (the plates of the rf tubes) on one side of the Heising choke, and 3500 volts (the B+ supply) on the other side of the Heising choke. So the Heising choke is delivering 3500 volts x 400 ma = 1400 watts of power to the rf output tubes on positive modulation peaks. On negative modulation peaks, the Heising choke has 0 volts on one side (the plates of the rf tubes), and 3500 volts on the other side... so there is 1400 watts flowing into the Heising choke (and being stored in its magnetic field). So during each audio cycle, the Heising choke delivers power to the rf tubes on the first half of the cycle, and gets "recharged" with energy stored in its magnetic field on the 2nd half of the modulation cycle.

That accounts for half of the remaining 2800 watts of peak rf power.

Now for the final piece, the remaining 1400 watts:

The Heising capacitor always has 3500 volts across it (except at low modulation frequencies). However, the current flowing through the Heising capacitor varies from +400mA on positive modulation peaks to -400mA on negative modulation peaks.

So... on positive modulation peaks, the Heising capacitor delivers 1400 watts to the rf tubes (the missing piece of the total) by drawing down the stored energy in its electric field (E= 1/2 CV**2). On negative modulation peaks, the energy stored in the Heising capacitor is replenished.

So, in summary... for the Heising configuration... where does the total of 5600 watts of rf input power on modulation peaks come from?

1400 watts from the B+ supply
1400 watts from the modulator
1400 watts by drawing down the energy stored in the Heising inductor
1400 watts by drawing down the energy stored in the Heising capacitor.

SDP

[KC2IFR]

Stu,
Thats the best description so far.....
I hope it helps Tom.......
It sure helped me and my simple mind


Bill

[K5UJ]

Tom if you think you are asking stupid questions, I'm the guy who didn't know what this meant at the start of the thread:

"triode-connected pair of 4X1’s."

I eventually managed to decode it and figure out we're converting tetrodes to triodes and 4X1 is shorthand (the X instead of a hyphen threw me).  So I'm learning a lot and trying to not get discouraged hi hi.

To confuse things, I attempted to answer your question for myself before looking at the answers.

I took 1000 w. times 50 ohms (feedline Z) and took the square root of that for the voltage and divided by 2 for the positive half of the cycle.  Figuring this was RMS I multiplied it by 1.414 and wound up with 158 v.  I figured the audio straight for dc as sqrt(500 * 50) and also got 158 v.  the sum of those two v. squared / 50 gave 2 kw.  I think for pep that is doubled? 

I have a heck of a time getting this also, and as many times as I have read all the math about modulation and pep I never can get it.   I like Brett's headroom suggestion about having modulation power equal to carrier.  Seems like a lot of the broadcast rigs have the same tube lineup in the modulator that they have in the RF.  I would like that especially for going 120 or more % positive.  I once calculated (somehow) that 130% positive is 5.2 times carrier.  I don't know for sure if that is correct but I use that in setting my carrier so as to not push my amp too hard.

73

Rob
K5UJ

[DMOD]

I have a heck of a time getting this also, and as many times as I have read all the math about modulation and pep I never can get it.   I like Brett's headroom suggestion about having modulation power equal to carrier.  Seems like a lot of the broadcast rigs have the same tube lineup in the modulator that they have in the RF.  I would like that especially for going 120 or more % positive.  I once calculated (somehow) that 130% positive is 5.2 times carrier.  I don't know for sure if that is correct but I use that in setting my carrier so as to not push my amp too

The total average power of each sideband in AM is Pc*(m^2/4). m is the modulation percentage.

The total average power of the carrier plus sidebands is Pt = Pc + Pusb + Plsb, Pc is carrier power, m^2 is modulation percentage squared, and Pusb or Plsb = Pc*(m^2/4)

so Pt = Pc + Pc*(m^2/4) + Pc*(m^2/4)

or combining terms:

Pt = Pc + Pc*[2*(m^2/4)] = Pc*[1+ (m^2/2)]

For example for 10% modulation with a Pc = 1000 Watts,

Pt10% = Pc*[1 + (0.1^2/2)] = Pc*[1 + (0.01/2)] = 1000 Watts*[1 + 0.005] = 1000*(1.005) = 1005 Watts

for 100% modulation with a Pc = 1,000 Watts, m = 1, Pt100% = Pc*[1+ 1/2) = 1000 Watts*[1.5] = 1,500 Watts

Using the formula for other modulation percentages:

Pt25% = 1031.25 Watts

Pt50% = 1125 Watts

Pt75% = 1281.25 Watts

Pt90% = 1405 Watts

Pt120% = 1720 Watts

Pt125% = 1781.25 Watts

Pt130% = 1845 Watts

Pt150% = 2125 Watts

[WD5JKO]

DMOD,

   Your calculations clearly illustrate the need to achieve the maximum percentage of modulation possible with the least amount of distortion. Equalizing that modulation is also important depending on a persons voice, and microphone.

   So those figures you show for lower modulation percentages clearly illustrate why some of us might transmit a big fat carrier across the globe, but with that carrier we are running QRP sideband power. That explains to me why I often hear strong carriers on 75M across the AM window here in Texas, but I cannot copy a single word due to the low sideband power being transmitted, and the receiving S/N ratio. Then other stations come in, from the same area, sometimes with the same or lower signal strength (carrier), and I understand every word spoken.

   So instead of modulating an 833 with a pair of 6AQ5's, can we modulate a 6AQ5 with a pair of 833's? 

Jim
WD5JKO

[N2DTS]

Its not just modulation percentage.
Its more the average modulation level on the carrier.

Many stations have a very low average modulation level.
I was listening on 80 yesterday and heard someone using what sounded like a very clean setup of some sort of modulated  solid state amplifier I think it was, something like an 8 ohm to whatever ohm transformer in the power feed to a solid state amp, or some crazy setup like that.
The audio sounded VERY good, which has to do with the voice as much as the transmitter, and the way the operator speaks. But as the operator talked about the setup, I noticed that the average amount of modulation was quite low, but with very high positive peaks. The other station commented on the unusual modulation showing on his scope, and it did look like 200% positive modulation, but only in very short duration spikes.

It sounded real clean, and very nice, but not loud, despite the 200% modulation.
I did not hear any sort of compressor or limiter described in the equipment lineup, a good mic, a preamp, a 3 band EQ, and some sort of audio power amp, no compressor/limiter.
I also noticed it got quite wide on peaks, no limiting of the extreme high end audio I assume.


I have talked with other guys who ran 75% modulation, and when I told them, they said it sounded cleaner that way and there was no need to run more....

With a normal rig, without any compression, the average modulation level will be quite low at 95% neg and 115% positive modulation.
With a strapping signal, this sounds good, but as the signal gets weak, it does get hard to copy.
Back away from the mic a little and the modulation level drops off severly.

If you are using vintage gear with a D104, you cant do much about it without changing things a lot, but at 100 watts, many guys just crank it up a bit and overmodulate a little in the neg direction.
With 100 watts or less, that is no big deal.

With all the modern gear avalable cheap, it is very easy to add some good compression, I use a beringer vx2000 that I got new for $100.00, it does limiting, compression, noise compander, voice freq EQ, and other things, plus has phantom power for good microphones, and a phaze reversal switch, a lot of bang for the buck.
In the past, I used an old DBX unit that compressed above an adjustable set point, and compressed at any ratio you want, from nothing to hard limit. That was a $10.00 hamfest buy.

One interesting thing I want to play with is the carrier adjustment on the flex, since I am running it into my amplifier with limited plate dissipation, I might try to lower the carrier and increase the modulation.
At 50% carrier, I should be able to get 200% positive modulation, 99% negitive, and reduce the amp load.
If I run 150 watts carrier and 1000 watts pep, that might sound quite loud with compression!
With it set at 90% I get 150 watts carrier andabout 800 watts pep.
It seems to sound loud in the mod monitor at 90.

The other thing I have been noticing is the WAY some guys talk on the air.
Some seem to assume some sort of 'personality' which I wont comment on, some sound VERY distracted, like talking to someone who is surfing the web or watching TV while talking. They often talk about themselves only, or their equipment, at very long length, dont seem interested in what anyone has to say, and spend lots of time adjusting things. Do they know they sound like this? Do they care? Are you just an excuse for them to listen to themselves talking on the air?

Sometimes it 'might be' a flex thing, since one operator I heard was using one, and sounded very odd, like I do if I try and monitor myself with its delay, like a bad echo on the telephone, you have   to   talk   sloooowwwwly,    and    sound     like   someone         with      brain            damage      .....

I am likely guilty of some of this, but I am aware of it and TRY to interact on the air like a real person.
I wont even bother to talk with some of these people, since a simple question about anything results in a 1/2 hour transmission about every piece of gear in the shack, with its history, cost, choices, etc.


I try to remember that it is nice to interact and talk with someone like you would in person, instead of addressing a convention or making a documentry for TV.

I dont mind long transmissions, its not that, its long transmissions to the audience about nothing that bug me.

If you catch me doing any of the above, please let me know, as the xyl says I DO tend to go off...

Brett

[K1JJ]

Stu,

Thanks so much for taking the time to write that up. It has answered my question - it does appear that for a standard plate modulated set-up, (non-Heising setup)  1/2 the peak power is needed from the modulator and 1/2 the power from the RF power supply. But what is surprising is your analysis about the "flywheel effect" from the inductor and capacitor when using Heising!  The integration/electromagnetic and differential/electrostatic effects certainly take a load off the modulator!   The break down of power contribution is an eye opener.   I suppose this is expected cuz the DC current is taken off the mod xfmr, but I've never seen it expressed in this particular way.

I've read it through a few times and it has added to my understanding of what's going on stress-wise in the system. BTW, I don't know if many of the newer guys (or farther away) know that Stu is/was a college professor of electrical engineering at a leading university.  That well-written and thought-out treatise shows it and should be put into the AM BB handbook.

While we're at it, Dave/APE, interesting to hear about the twin 10KW iron needing a small crane to move it... :-)  If I cud snap my fingers, I wud have its brother back here again.  

QUESTION: Dave, you mentioned using the Heising coupling cap in the ground lead of the mod xfmr. I realize that is used to reduce the HV insulation to ground requirements.   I already have all the mod related iron sitting on a Plexiglass sheet isolated from ground. Is there a STILL another reason for me to place the cap in the ground lead?


Good comments about keeping our  % of modulation high. There’s been a lot of threads about this in the past. The bottom line is to keep a hawk-eye on the scope and be sure you are constantly approaching negative 100% modulation most of the time. There’s many ways to maintain a high level. But the most important thing is to be sure your transmitter is capable of a clean 120-130% positive modulation. This can be determined by sweeping it with an audio gen. There is a short article in the East Coast section about doing this.


Brett: Your comments about some ops who talk to themselves are a scream…   The way to determine if we are guilty of it is to tape ourselves when on the air. I have some tapes of myself from 20 years ago and I cringe when I hear some of them. I made some old buzzard 10 minute transmissions that just rambled on about useless crap. I was bored to death listening to them a few months ago. Over the last 10 years I’ve made it a rule that if I’m gonna get on the air, it’s gonna be an “event” and I will have something to add or material to entertain with. There’s got to be an agenda of some kind or we all end up just hanging out and boring ourselves. It takes some thought to keep things interesting and we can’t let others pull all the load…  

Well, back to wiring up the new rig. The shack is a disaster zone. I’ve decided to rebuild most of the rigs here, some with major changes, so still have a lot of work left.

T

[w3jn]

It's always a special event when you get on, Tommy.  I can honestly say I've never had a boring QSO with you 

[W2PFY]

Some seem to assume some sort of 'personality' which I wont comment on, some sound VERY distracted, like talking to someone who is surfing the web or watching TV while talking. They often talk about themselves only, or their equipment, at very long length, don't seem interested in what anyone has to say, and spend lots of time adjusting things. Do they know they sound like this? Do they care? Are you just an excuse for them to listen to themselves talking on the air?

Boy you nailed me on that one Brett! Sounds like how I operate. I didn't think anyone would notice.

Great dissertation Stu, I always learn from you

Tom, it seems you never can stop building and experimenting. Me old hat is off to you

[N2DTS]

Well, I think some of what people talk about on the air is booring to someone, that is not the big point, and you cant help that, but when someone only talks about their stuff and does not interact, or ask questions about the other guys stuff, or talk vintage gear, car stuff, some crazy homebrew project plans, etc, it gets old very fast.
Talking one way, and taking no interest, or attaching no importance to the people you are talking at is just plain rude.
So is the attitude of "my stuff is great/and or expensive and nothing you have or do interests me at all".

Another thing that seems rude to me is chatting over top of someone, is anything more rude?
Where does that come from?

I have enjoyed some of the qso's lately about hot rod cars, guns, Tim-trons yard cleanup, etc.
I always love hearing about the guy running a screen modulated 4cx250b, or water cooled 6aq5's, and so on, sdr is always interesting, and vintage gear is interesting to hear about, the really good sounding dx60B is very interesting, antenna's and tuners, even current news items.

Brett

 

[W2PFY]

Back to the topic. Where is the best place to have the hysing cap placed? Some people put it to ground or common at the so called cold end of the mod transformer. Others like me in the case of the Westinghouse MW-2 have it from the so called hot end of the mod transformer to the hot end of the mod reactor. I have the B+ to the bottom of the reactor and mod transformer tied together. My thinking was that I wanted to avoid current surges going through the secondary of the mod transformer every time I keyed the transmitter. This doesn't happen all that often on a broadcast transmitter because they are always on. I was also lead to believe that putting the cap to ground gave additional dc filtering. So to get all that energy stored in the reactor and audio coupling capacitor, which way is best? I've seen it done both ways by manufactures, so is it the same? Seems to me the cap to ground would make more sense even if I don't do it that way.         

[AB2EZ]

Terry

If you put the cap between the transformer and ground, then you will have the transformer's secondary winding "floating" at the B+ voltage. This increases the voltage between the secondary winding and the case (assuming the case of the transformer is grounded). On the other hand, it reduces the DC voltage between the secondary winding of the transformer and the primary winding of the transformer (assuming the center tap of the primary winding of the transformer is connected to B+).

As a compromise, I've seen some recommendations to place twice the desired Heising capacitance on either side of the secondary (one to ground, and the other to the Heising choke). You need to include balancing resistors to split the DC voltage equally.

Another compromise is to place the capacitor between the secondary and ground... and to float the transformer's case on appropriate insulators... with rubber shock absorbing grommets to dampen "talk back" while you are at it. In this approach, one has to be concerned about the danger of the B+ appearing on the case of the transformer if a short occurs (or even a very small amount of leakage from the transformer's secondary or primary winding to the case)

In my KW-1, I have the mod transformer sitting on a "mouse pad" to dampen the talk back. Gravity, of course, keeps the transformer on the bottom of the cabinet. Opening the cabinet door when the B+ power supply is on, and poking my hand around where the mod transformer and the plate transformer are located, is not a good idea in any event... so I decided that floating the KW-1's modulation transformer's case is okay.  

Best regards (and thanks to all for the kind words!)
Stu

[N2DTS]

When I was playing with this stuff, I think it was Don C who told me which way was better and why, but I dont remember!
I tried it both ways and could tell no difference.
What I DID find was no matter what value of cap i used, it always sounded different.
Small, med, large, really large, each one had a different 'sound'.

That was the main reason I scraped it, I could not figure out what I liked best...

Brett

[K1JJ]

Stu,

OK on placing the Heising cap in various positions and the compromises.  Maybe the two-cap approach is the best. Or, as I intend to do, float the case above ground and put the cap in the secondary to ground.

Another question:  In the distant past I used a large value Heising cap, like 10-20uf. But I was told it should be a small value, like about 2uf. I have since used 2uf.

What is the reason for the smaller value?  (If you agree with this idea)

T

[W2NBC]

Tom,

I found Stu's detailed description in this thread to be extremely helpful regarding the Heising cap.

http://amfone.net/Amforum/index.php?topic=18722.0



Jeff