Need a BOOST? A slick way to get another HV supply tap

[K1JJ]

My HV power supply is set up to give 2KV and 3KV.  The other wiring alternative would be 4KV and 6KV which is too high.  I needed 4KV to power a linear amplifier.  I tried every configuration involving full wave, choke input, etc. but no luck.  I didn't want to use a Variac due to sag.

Here's the solution:  A strapping transformer booster in the 240V  primary.  

I use two identical HV transformers that have the primaries in series  and the secondaries in parallel. The result is 120AC across each primary using a 240V source.  The transformers have 240V windings so I could add more voltage in series with their primaries without a problem.

I used an old 100 pound Variac core and wound 70 turns of #6 insulated wire on it. This gave me about 75V out with 240V on the Variac primary. IE, I converted a Variac autotransformer into a pri/sec transformer.  This extra 75V gave me the 4KV I needed out of the HV supply. I used a pair of breakers to turn the booster on or short it out when not used.  At 75V, it only has to handle about 1/3 the total power of the supply.

I ran a current test on the HV supply and found the supply was still very stiff when the booster was in line.  So now I have 2KV, 3KV and 4KV available to power the various rigs. The filter cap is rated at 100 ufd @ 10KV, which helps...

Look at the picture below and notice the cigarette lighter for scale. It really is a big MOFO booster and will handle about 100A in the secondary.  Maybe someone here can use the idea when out of ideas - like I was.

BTW, this is the same booster that got pulled and rammed against the filter choke when the HV supply shorted out. The filter choke acted like an electromagnet. I've seen this happen with big mod iron too. It pays to tie down the iron for this reason when in close proximity.

73,

Tom, K1JJ

The 100 pound Booster:

[flintstone mop]

Hey Tom
That's quite a booster!!
I used to work on TV's and they had a boost circuit in them to get the 160vdc to 700vdc for screen volts on the CRT. Would that be of any use? Probably current would be a serious limitation......dunno

Fred

[WBear2GCR]

Ya lost me on the curve there Tom!

You say your booster gives an output of 75vac with 240vac on the primary??
That's less not more voltage?

Or, what is your AC voltage input to the primaries of those HV transformers *after* you apply the booster??

Howz about a schematic?

I don't see how this works, or where it goes... etc... but I am slow anyhow.

                               _-_-bear

[Ed/KB1HYS]

Bear,

he's using the variac transformer as a buck-boost transformer.  The 75V out gets added to the 240vac for 315vac input to the primary of the plate tranny. 

[K1JJ]

Yes, Ed, you have it correct.  Actually, I see the schematic says 68V is added to the 240AC primary giving 308V total when the boost is on.   The sec of the booster transformer MUST be shorted when not in use or else some of the 240 will drop across the booster xfmr. I simply use a breaker for the job.

Here's a handwritten schematic to give an idea:

T

[k4kyv]

I have found that variacs have an unacceptable sag under full load.  One of my biggest mistakes ever was giving away the tapped kilowatt size 120 v.a.c. auto transformer with built in rotary switch that came with an old x-ray machine I salvaged from the dump. That thing was as stiff as the a.c. mains voltage feeding it.  I couldn't measure any sag at all. I replaced it with a neater and more compact 20A G-R variac, but was never happy with its performance.

[KE6DF]

I have a couple Signal Transformer step down transformers.

One of them has two 28V CT windings that are intended to be connected in series or parallel.

I was thinking that would make good buck/boost transformers for a high voltage supply.

By connecting the two secondaries in series, it would give you a choice of 4 taps each 14 volts apart.-- plus the direct line connection.

Connecting those up with a high current rotary switch would give a nice variable HV power supply.

My transformers are rated at 340 and 425 VA so would only work for a moderate output HV supply.

But Signal has larger sized similar transformers available.

I'm not sure if using one of these would be any better in terms of sag than using a variac, though.

But it's another idea, FWIW.

Then there is the old Thyratron phase angle variable supply method commonly used on industrial RF applications like inductive heaters...

[Bill, KD0HG]

Thyratrons? That's an interesting concept...Why couldn't one use a heavy-duty light dimmer gadget on a plate supply to adjust its voltage? Use a couple of 50A SCRs or Triacs?
Same effect, no?

I believe that's how Continental adjusts the screen voltage on their 30+ KW FM transmitters.

[K1JJ]

I agree, Don -  the Variacs I've used, even very large ones sagged under full load. After all, they have to pass the same power as the HV transformer, so should be at least as big.  

The advantage of the boost transformer is that it only has to pass the power represented by the extra voltage it supplies.   So if the circuit is say, 10A, the booster will still supply 10A since it is in series with the main xfmr, but it supplies only a part of the voltage. If it supplies 70V out of 308V, then the main xfmr supplies 2400w and the booster supplies 800w.  I like that vs: a Variac having to handle it all.

Variacs are great for low power supplies like grid, screen, exciter plates, etc, but for high power HV supplies, not.  But one alternative is to use one to slowly ramp up the HV and then set the Variac so it is basically out of the circuit when the tap hits the top of the winding, making a direct connection. But then, the value of a variable variac is lost.

Dave/IYH - Yes, those low voltage transformers that handle lots of amps (like fil xfmrs) work FB.  A pair shud serve you well, esp bucking for more versatility.

T

[KE6DF]

"Use a couple of 50A SCRs or Triacs?
Same effect, no?"

An SCR is pretty much the solid state equivalent of a thyratron, I believe.

There are thyratron equivalents of all the common MV rectifiers.

A 5557 is a gridded 866, an 873 is a gridded 872a, and there are some that are gridded versions of 575A's also but I forget the numbers.

There is a circuit for this in some of the old "Radio Handbooks" in the power supplies section.

I wonder how well it would work if you did the phase angle "chopping" in the primary circuit?? Perhaps just fine.

Also, I'm not sure how high voltage ratings go on SCRs. If you can get them in the 10KV range, you could use them as rectifiers in a HV supply.

[Bill, KD0HG]

"Use a couple of 50A SCRs or Triacs?
Same effect, no?"

An SCR is pretty much the solid state equivalent of a thyratron, I believe.

There are thyratron equivalents of all the common MV rectifiers.

A 5557 is a gridded 866, an 873 is a gridded 872a, and there are some that are gridded versions of 575A's also but I forget the numbers.

There is a circuit for this in some of the old "Radio Handbooks" in the power supplies section.

I wonder how well it would work if you did the phase angle "chopping" in the primary circuit?? Perhaps just fine.

Also, I'm not sure how high voltage ratings go on SCRs. If you can get them in the 10KV range, you could use them as rectifiers in a HV supply.

The days of large power contactors in radio transmitters seems to be over...

I recently installed a 50 KW broadcast rig and it's all SCR controlled. Totally quiet and a slow ramp up at turn-on. Primary voltage is 480, 3-phase, fused at 200 amps, the power transformer weighs in at around 700 #.

I don't miss those giant relays with dime-sized contacts at all, and they sure do make some big-time huge SCRs.

[KB2WIG]

" I don't miss those giant relays with dime-sized contacts at all, and they sure do make some big-time huge SCRs "

I still pine for the days of rotary equipment,  knife switches and their bright green arc.

klc

[Tim WA1HnyLR]

enter another method of obtaining an intermediate HV outpoot. One of the WBCQ transmitters uses two 7200 volt 25Kva pole pigs for it's HV supply. Primaries and secondaries paralleled up . Full wave bridge /ckoke inpoot delivered about 6Kv under load. This proved to be a bit much voltage for the PA (6 -3cx2500A3 tubes in push pull parallel ) There was a need to drop the voltage. You could series up the primaries which would halve the outpoot voltage . The transmitter would not make power this way with 3Kv on the PA tubes. I decided to try something a bit different. I left the HV secondaries in parallel and had the primaries  in series BUT, I connected the 240VAC to the center tap of one of the pole pigs.while the other transformer had the full primary utilized. This resulted in an outpoot voltage of 3800 volts. This setup has been working well for years now. Tim WA1HnyLR

[VE3GZB]

I like how you've made it possible to switch in and out the choke from the HV DC side. Nice!

[WA1GFZ]

SCRs generate random broadband swithing noise that is a pita to filter at high power.

[Bill, KD0HG]

SCRs generate random broadband swithing noise that is a pita to filter at high power.

But they're on only when you're transmitting.

[KC2IFR]

Hi Tom,
Looks like a buck/boost transformer used frequently in apps that need further voltage adjusting.

[K1JJ]

Hi Tom,
Looks like a buck/boost transformer used frequently in apps that need further voltage adjusting.

Yep, Bill, that's exactly what it is. In my case, I needed an extra 1000 volts to run the linear and had no other way to do it without buying another HV transformer.   Having a switchable 2KV, 3KV  and 4KV is just what I needed for all the rigs.

T

[KE6DF]

One other option to get two voltage choices out of a single supply when you are using a choke input filter is to add an input capacitor right before the choke.

If you switch the input cap into the circuit, you have a capacitor input filter. When you switch the input capacitor out of the circuit, you have a choke input filter.

So if you have, say a 2KV RMS tranformer, with the input cap switched out, you would get about 1750vdc. If you switch the input capacitor into the circuit, you would get 2500VDC or so.

When the input capacitor is switched out of the circuit, you could connect a high resistance between one side of the input capacitor and ground (the other side stays connected to the input terminal of the choke).

When then you switch to high voltage mode (by shorting across the resistor), the cap is already charged up part way to the DC output value and the current surge on the switch would be lower.

FWIW, just one more random idea that wasn't mentioned.

[K1JJ]

Yep, that is a good idea.

I'm already using it - look at the schematic I posted back a ways.  By shorting out the choke (in the CT of the bridge diode stack) it becomes a cap input. That's how I go from 2KV to 3KV under load.  Then I add the boost to go to 4KV...

T

[Gito]

Hi Tom

Can this circuit adapted to the for power supply ? the schematic that You have Attached,it also used a booster transformer,and like you this transformer  only supply the needed wattage of the 75 ac output.


Gito

[K1JJ]

Hi Tom

Can this circuit adapted to the for power supply ? the schematic that You have Attached,it also used a booster transformer,and like you this transformer  only supply the needed wattage of the 75 ac output.


Gito

Hi Gito,

Sure, it would work fine.

It is a different way of conecting it and is clever....   Thanks for the suggestion.


T

[Gito]

Hi

When using this circuit don't forget the polarity /phasing of the transformer.
Because it can boost or buck the output AC voltage.
here's a picture as an example

Gito

[WA1GFZ]

That core looks like it is from 30 amp 240 volt variac.
I have one in my rig and never noticed sag but only use it for tuning up then it is set to max.

[KM1H]

How do you justify running 315V on a 240V winding? Have you measured the primary current in idle and full load?

Carl
KM1H