Power Supply... MOT style.

[Ed/KB1HYS]

I had built a full wave rectified B+ supply for my HB rig.  I used two Microwave Transformers parallel-primary, series secondary worked FB, tested and would put out ~ 0.5 amps at 2200 volts all day long with out breaking a sweat (it did hum a bit while it worked, but then so do I  ).

However, while both tranies looked identical they were not.  One had ALUMINUM wire on the secondary! (cheap!)  It still worked ok, but the only connections I could make were crimp type, I wasn't happy with how that worked out.  While I was trying to make a better mechanical connection I broke the wire off, right at the bobbin!  (GONE). Aluminum seems to have an issue with fatigue .  Thus rendering this layout NFG... 

So... 

I added two more rectifier tubes for a full wave set-up, (added two additional filament windings to my HB fillament transformer with extra insulation too).  I have a huge beast of a transformer that I pulled out of a commerical MW that was at the dump (2kW unit!).  I'll drop that in there with a booster (the alum 2ndry one  with the secondary chiselled off and rewound with COPPER as a LV 60-80V) and get about 2800 (or more!) volts out!   I thought about using a couple of taps and a switch on the booster to have a high/low power out setup, but decided to just go for the gusto, this is going to be a QRO rig so, no messing about with low power stuff.

Initial tests look like this is going to be more robust than the twins in full wave. I'll have to put the salt-water load back in service and really work it hard to see if it'll survive.

You could easy make a plate supply for a pair of 813s with these guys, they have about the right voltage  "out of the box", plus they come with diodes that are rated for them too. 

[WA2IXP]

Ed,
  I have had good luck with these transformers also.  20 or 30 turns were added to the primary in space provided by removal of the shunts and filament winding. For single transformer use I selected the ones with the most favorable inner winding end to core insulation. So far none have shorted to ground. Have also cut or milled open the cores and added air gaps to use as chokes. What are the details or your salt water load??
                                                                                                                 jay-

[K9ACT]

I knew there was some reason I wanted to look you up after our QSO last week.

As discussed, I added 85 turns to the primary and you motivated me to remove some to see how much poop I could get before saturation set in.

I did it 5 turns at a time and only got to 10 before the line in increased by 1 amp with no increase in output.

I suppose in the name of science I should keep going but I need more moral support.  There is a big difference between 75 and 25 so maybe all MOT's are not equal?

I am now running at 1800 VDC with a .3 amp load.  I have no way of loading it any more and I only need .220 for the 810.  It looks like it would easily handle 2 tubes.

I have attached a picture of the supply.  Although it's not very pretty, it is important to note that the pw trans is much smaller than the filter choke.  This whole thing weighs about 20 lbs.  It is fun to compare it to the supply in my 2 x 8000 rig weighing in at about 90 lbs with the Dahl tranny.

The next step is two transformers but I need a reason (new project) to do it.

n.b. The label on the front is data from the transformer replaced by the MOT.

js

[WA2IXP]

John,
  I could have a few more primary turns than I remember and yes, the unloaded primary current varies from unit to unit. I have checked 6 or 7 and found the current between 2 and 4 amps. The one I used wasn't real high to begin with and the extra turns were to bring down the voltage from 2400 to about 2100. It is about 1980v dc with .3 amp load. It has 47 mf of filter. The inrush current is controlled with a resistor switched with a solid state relay. A similar relay controls the line input switching. Delay on the resistor shorting relay is acomplished with a series resistor and an electrlytic for the 24vdc control voltage. I am interested in Ed's load resistor. I'd like to have a 5 or 600 watt load to test hv power supplies but only have wirewounds up to about 200 watts.
                                                                                    Jay-
                                                                               

[Ed/KB1HYS]

BE CAREFULL!!!! HIGH VOLTAGE CAN KILL YOU DEAD period. POWER OFF BEFORE DOING ANY ADJUSTMENTS AND USE A Grounded PROBE TO MAKE SURE EVERYTHING IS OFF AND DISCHARGED.  It is better to replace a component or bit of wire than to have your wife or kids find your blackened carcass over a rig.

The salt water load is simple and dates from the beginning of the 20th century when it was a wooden barrel and iron plates used to test the output of big reciprocating steam powered electric generators.

 I use a large glass jar with a plastic lid ( I initially used a BIG pickle jar). Drill a three of holes in the lid, two are for a wire and a piece of brass threaded rod sized for a snug fit.  The third is very small and just a vent.  I put the rod in the center and the wire and vent near the edge.  In the center hole Use a short (2-6inch) length of threaded brass rod (size the hole to enable the threads to grip this helps with adjustments later).  Put a good solid wire through the second hole so an inch or two is under the water, apply some goop (RTV or something) to hold it steady.  It's VERY IMPORTANT that the two leads can NOT TOUCH. for obvious reasons.

Fill the jar with CLEAN water and screw the brass T-rod in until a short bit is contacting the water and measure the resistance -  should be fairly high with  just clear water (but I bet it will be lower than you expect!).  Turn the threaded rod down and recheck the resistance.  If you can't get the load value you need, add TINY amounts of salt to the water and stir (with the City Water here, I added a little under a half tsp of salt to get a 4000 ohm load- YMMV).  Measure the resistance often as it's easy to go over and have too low a resistance. If that happens you'll have to mess around with adding clean water or just start over.

When you get the resistance value you want,  securely connect the leads from the power supply and apply power to the load.  If you use a Good size jar it can take quiet a bit of power before it gets warm.  There is some adjustment using the rod so one load can have different resistance values for different tests. 

Once made it will last forever unless you leave it out in the shack and it freezes! (like mine).

Also, use similar metals, else you could wind up making an electrolytic rectifier instead.  I did this using an aluminum and some other alloy electrodes and got a really high resistance one way and a VERY low resistance the other. Found out later that this was a technique used back in the 20's to form HV rectifiers.  Cool, but not what I wanted.

BE CAREFULL NOT DEAD.

[Opcom]

for those nasty aluminum wires, there is a product called alumi-weld. it supposedly will bond aluminum to other metals. I used it, or something like it, to repair a variac that had been dropped. The only issue is the temperature needed is higher than solder. A heatsink pliers clamped to the wire next to the winding could help with the heat.