[Amps] MW Oven Transformer question

[Will Matney]

Tony,

To be honest, I forgot about those welds. If the welds are close to the ends of the core it wont be so bad as the flux will have already started to bend down into the legs. What the welds do is act as a thick piece of iron would if placed where the flux will pass through it. Also, it's according how deep the welds are, but they will increase the losses if over in the magnetic path. A thick piece of iron has high eddy currents and is the reason for using thin laminations. The thinner the better and it increases with frequency so you have to use thinner lams as you go up in frequency. In these, I'd say the lams are either 14 mils or could be a shade thicker. They'll try to use the thickest they can to cut down on the time it takes to stack a core. They're penny pinchers and this is evident. Really, the thru bolts holding the lams together should be insulated if they're steel because of this same thing as they can heat up but another reason exsists, read on. If they're placed way up in the corner as most are, the effect isn't as bad for the same reason that the flux has bent down the leg and missed them some. They'll still get some, but not as much if they were down in the leg itself. An uninsulated screw acts as a short between the lams and will have high eddy currents in it plus can act sort of like a 1/4 turn winding if it's grounded to the frame! If you have a low voltage in a winding, what do we get with a short, very high current. The lams are insulated between each other to lower the eddys. A weld counterdicts this where it's at. Most of the screws will have an insulated step washer on each end and some have used thin tubing over the whole screw. This being said, and according to how many welds on it, I'd knock the efficiency back to about 75-80%. It would be a good bit lower than this for one running by itself. By the flux density being so low, the efficiency will be high.

Best,

Will


*********** REPLY SEPARATOR  ***********

On 3/21/06 at 5:03 PM Tony King - W4ZT wrote:

>Will,
>
>Do you think that the welded laminations will not be a problem and 
>increase losses? Beyond that, what you're saying is good news.
>
>73, Tony W4ZT
>
>Will Matney wrote:
>> Gerald,
>> 
>> My guess by seriesing two will drop the flux density in each to around
>10 kilogauss as each will have 1/2 the line voltage across it. That's well
>below the bottom of the knee and in the linear region. It will raise the
>power capability X2 also, but I can't say how much without knowing the
>core dimensions. Two this way though will run as good or better than a lot
>of off the shelf linear transformers because of the low flux density. This
>means the efficiency is really high, probably 95% for a rough guess.
>> 
>> Best,
>> 
>> Will
>> 
>> 
>> *********** REPLY SEPARATOR  ***********
>> 
>> On 3/21/06 at 2:22 PM TexasRF at aol.com wrote:
>> 
>>> Thanks Will and John; I now have an understanding of the heating issues 
>>> with 
>>> the MOT. It appears that these transformers are really a poor choice
>for  
>>> continuous operation such as a filament power source. Two transformers
>>> with the  
>>> primary and secondary windings series connected will work and for 
>testing
>>> the 
>>> traveling wave tubes is the quick method to be used.
>>>
>>> Many thanks for your help in my continuing education!
>>>
>>> 73,
>>> Gerald K5GW
>>>
>>>
>>>
>>> In a message dated 3/21/2006 11:10:15 A.M. Central Standard Time,  
>>> craxd1 at verizon.net writes:
>>>
>>> If you  want to add turns to the primary (if there's enough room, and
>>> should 
>>> be with  the secondary removed), you can do a simple test to see how
>many 
>>> turns to add.  First, use the formulas I published last week about
>>> figuring the 
>>> core size and  number of turns to find the correct number of turns for
>the
>>> core 
>>> size you  have. Next, connect the primary to 120 Vac and wind a 10 turn
>>> coil 
>>> secondary.  Take that secondary voltage and divide it by the 10 turns.
>>> That 
>>> will be the  turns per volt. Last, wind the extra turns you've found you
>>> need by 
>>> the  formulas and the known turns per volt. One good thing on this
>>> transformer 
>>> in  question, the primary is on the inside. If on the outside, you'd be 
>>> screwed on  removing the secondary.
>>>
>>> Best,
>>>
>>> Will
>>>
>>> *********** REPLY  SEPARATOR  ***********
>>>
>>> On 3/21/06 at 11:42 AM John Popelish  wrote:
>>>
>>>> TexasRF at aol.com wrote:
>>>>
>>>>> Not what I hoped  to hear but I appreciate the good info! 
>>>> If you have a supply  of this kind of landfill, you might put two 
>>>> similar units side by  side, wire the primaries in series (effectively 
>>>> halving the volts per  turn, so eliminating the saturation problems). 
>>>> Then you can knock the  shunts out and wind a similar secondary on
>each 
>>>> (doubling the number  of turns to compensate for the half primary 
>>>> voltage) and wire those  two secondaries in parallel.  This gives you
>a 
>>>> reasonably  efficient transformer (that won't overheat without a fan) 
>>>> with about a  kVA rating.
>_______________________________________________
>Amps mailing list
>Amps at contesting.com
>http://lists.contesting.com/mailman/listinfo/amps