The AM Forum

I would imagine that this has been beat to death in the past BUT,
does anybody know what Antek transformers were used for the ART-13 power supply.
Somebody on here used them but didn't say what models they used.
Thanks,
Don W9BHI

I've  built 3 power supplies for the ART13. Looking at that site I've come up with this arrangement. Using these toroidal transformers you should be able to make a nice compact power supply.

AS-4218          18V   11.1A          25.452vdc   $44.00
AS-2T300        300V   0.33A          424.2vdc     $38.50
AS-4T500        500V   0.4A x2       1414vdc       $57.00

These voltages are through a full wave bridge

73 de Gary,KF9CM
www.kf9cm.com

  Gary,

   This is good info. I wonder though about those higher voltage Anetek transformers. If you series the secondary windings as you suggest here, [500+500] * 1.414 = 1414vdc (FWB, cap input filter), is there an insulation issue between the windings that we need to worry about? I presume these windings are bifiliar wound, so when we series these up the high side of one winding needs to connect to the low side of the other. Looking at the Anetek web page, I see no mention of concern, or any text describing how those secondaries are wound. I do recall somewhere, possibly by Stu (AB2EZ) raising this issue.

   If the series windings is a concern, then possibly leaving them parallel, and instead use the dual diode full wave doubler scheme would be a good alternative, and should work just as well..possibly better since the winding resistance would be 1/4th that when the windings are in series.

Jim
WD5JKO

Hi Jim,

Anetek Hi-pots the AS-4T500 to +3500 volts, which the 1414vdc is well below that rating, still I would use Teflon sheet between the transformer and the chassis. If 1414 volts seems a little high you can opt to a lower voltage transformer. I've heard of guys using up to 2200 volts on the 813s plate by standing off the modulation transformer from the chassis. 


73 de Gary, KF9CM
www.kf9cm.com

Gary,

   I tried to download the AS-4T500 PDF but could not. I could download the AN-4T500, and here is what it says:

"The dielectric test is more than 3500V in between primary and secondary coils."

This says nothing about dialectic strength between the secondary coils, hence my concern and question. Also I've seen it said before that for a given chunk of iron (transformer), the full wave doubler is more efficient then a full wave bridge. This assumes a capacitor input filter. The main reason is less copper is needed (1/2 the turns), and hence less resistive I^2*R copper loss. With the Anetek chosen, using FW doubler with the two 500v windings in parallel (instead of series) will provide the same voltage with 1/4th the loss from secondary winding DC resistance. Either way, it will work so long as there is no internal breakdown between the two bifiliar wound secondary windings.

Jim
WD5JKO

Gary,
This is just what I needed.
BTW if using the 1.414 calculation, wouldn't 20 VAC get you closer to 28 volts? (28.28)
When it is loaded it should drop a little bit?

Thanks very much for the info.
Don W9BHI

I just looked at W1ID's post and realized the transformer is spec'd at 115 volt's.
My bad :P

Don W9BHI

good for you, Jim ... I raised this issue of construction technique of the transformers in the discussion with Stu .... I disassembled one of the secondaries of an Antek step down transformer to see and indeed the secondary was bifilar wound ... this still leaves the question open for a step up transformer .... indirectly tested with a leakage tester and when using a NE2 as a low current detector winding end to other winding start and no other connection started seeing leakage at 1200 V or so ... this could be 2 formvar layers or capacative coupling .... since the test was DC that lets out capacative coupling ... I would only use the 500V+ windings in parallel ... I believe the lv to hv windings insulation to be ok at 3500 V but not actually hipot tested ...73 ...John

I got email back from Ango Technology Corp.(Antek) and they said that you could series the 500 volt windings to get 1000 volts on the AS-4T500 with no problems.  Why would they sell a transformer that would break down in that configuration.

hi Gary .... at least you got it from the horse's mouth .... John Ango must have been busier when I asked that question 18 or so months ago .... I wonder just how far far he will warrant them or explain his breakdown tests a bit more ...73 ...John

Hi Don,
   I have an ART-13 in need of a power supply. Please let us know if you go with the Antek torroids and how they work out.

Thanks,
Ed  KJ4JST

Ed,
I just got my Meanwell 27 volt 13 amp supply and I am ready to order the Antek toroids.  
I will keep you posted on my progress.

UPDATE:

I ordered the toroids and asked to speak to an engineer about a suitable inrush limiter, his answer was (with a heavy asian accent) "I dont know".
So don't expect any technical help from these people.

Don W9BHI    

I got the transformers from Antek today,
Should I put the 500 volt secondaries in series and run a full wave bridge or put them in parallel
and use a voltage doubler to get the 1414vdc?
The two secondaries are rated at 500v@400Ma.
Thanks,
Don W9BHI

I believe the more conservative approach is to parallel the windings and use a full wave V dblr

I would also use step start or other I limiting

Nice iron and no problem either way at these voltages with separate transformers. I actually run a bridge and 100 uF cap stack with 100K equalizing resistors on my ART-13 and the hum is low enough that nobody comments. For religion - add a choke and another cap stack for Pi filtering. A variac is nice in the input because it can actually give you a bit of a boost in voltage and meters are sweet as well. Safety - lamps and fuse. I have a delay circuit with a relay that allows the HV to be able to be turned on 30 seconds after fils.

Hi WU2D,
   Are you using the 4T500 Antek torroid? Also, the 100 uF that you mention, is this the capacitance after the loss from stacking? What caps are you using and how many?

Thanks,
Ed KJ4JST

Series (bridge rectifier stated!) the Antek 500V secondary's = 1000VAC
[RMS x 1.4 (peak) = 1400vdc]  Insulateing the torroids from the chassis and
each other would be a good idea.  I use squares of plexiglass with a single
nut/bolt in the center.  In this case, three squares would be called for.

On the caps, that would be a stack of 4, readily available 450mf@450vdc =
110mf @ 1800V with four 100k resisters, one on each cap ...

The 100mf stated in the post was a rounded number, probably...

This is just a WAG on my part from READING WU2D's post...

Geez, why are you guys making this project so complicated?  It's
just a PS for an ART-13, simple enough.  

Hi Ralph,
   Thanks for your input. What you describe is what I first thought I would do. But then I thought I remembered hearing that a much higher safety margin  than 1800 v cap rating vis a vis 1400 v plate supply was needed; especially in a power supply for a plate modulated final. I envisioned needing 7-8 450v caps.
   Maybe I'm being overly cautious.

Ed

Capacitors see a big voltage surge during starting depending on the circuit.

Just for the heck of it, I ran the Duncan amps simulator on a power supply with the following design:

Solid state rectifiers.
Full wave center tap hookup.
Choke input.
10 hy choke, 50 mfd capacitor.
5K Ohm resistive load.

Transformer voltage 3KV center tapped:

Results:

output voltage: 1320 vdc

Peak voltage occured at about 0.1 seconds after turn on.

Peak voltage across the cap: 2090.

So I'd say, for this supply, better use a cap with 2500V ratings or higher.

Dave

Okay, here is some additional dope...

Those 450 volt caps have a peak rating of 525 volts generally, so that
takes the peak stand-off to 2100 volts, enough safety factor for the
job...

A simple input voltage spike control to take care of the start surge  (for
115 vac input voltage) is a 10 ohm, 10 watt resistor placed on a 115 volt
ac SPST relay wired so the relay shorts the resistor on closing.  The
mechanical delay is enough to kill the surge on the cap bank and things work
fine.

If you wish to go 1st class, there is a surge delay kit available from Jeff at
Harbach Electronics, (SS-100 Soft Start, $30.00 shipped)... 

In addition, that is if you take advantage of all the torriods listed by Gary,
KF9CM, IE, the 18, 300 and two 500 volt units, by putting the HV primary's
in series and the other two in series you can use a 230 vac input for the PS...

This will produce better regulation as well.  The surge resistor would then be
20 ohms, 10 watts.  

Thanks again Ralph. Very useful information. I'll start building this weekend.

Ed

One final word...  If you are uncomfortable with the peak voltage rateing
with only four caps in the filter on the HV output you can add an additional
450mf/450v cap & 100k resistor.  This will drop the total cap size on
the HV output to 90mf @ 2250VDC RMS, 3150V peak...  Personally I find
that a bit of overkill for this application (ART-13 Tx).

I have used 5 of these caps at 2450VDC RMS  with a swinging input choke
with no problems using the simple input spike (relay/resistor) system.  I also
have variable input on the AC but have been known to kick the HV supply's
with full output voltage on the HV transformers...   :o ::) ;D

Yeah, my introduction to "PLASMA 101" with a critters nest of dry
leaves outside the window of the shack where the open wire feeders
are...   Stupid field mouse figured that protected window sill was a
perfect spot, I guess...  

Hey, I'm not the first or only one on AM that has had a plasma fire on
a feed line...  Of course, if I were to run the 440 without the flame
thrower add-on, such hazards would not exist.