Tomm,
They're the same formula published in every transformer book I have including
the "Bible" so to speak named Practical Transformer Design, and one written by
M.I.T. back in the 50's. I have probably 10-12 books on the subject and all use
the same thing. There's some short ones discussed but only get you in the
ballpark. The long one I spoke of where you take everything into account like
paper type, and thickness, even the varnish thickness on the wire, but it still
is off in the real world I found. Not buy a big degree but enough that it's not
right. There's just too many variables involved like how tight the wire is
wound on each coil, the fish paper composition can change between batches,
heck, a whole bunch of stuff including the coil shrinking over temperature
variations. Some of this, any of the published formulas dont take into
consideration, and really I couldn't see how some could even be included. One
would think that a fudge factor could be found to add on where you
would really figure it to where you would actually be cooler than what you
come up with by a calculation. That is really what is needed, a safety factor
to where a temp quote will either meet the quote, or be under it. Then the
designer and customer are both happy.....
Best,
Will
*********** REPLY SEPARATOR ***********
On 1/29/06 at 5:07 AM KD7QAE wrote:
>I agree that all the formulas are estimations. If you know the hot spot
>temp at your testing lab's ambient temperature, you can simply add the
>difference between your customer's spec temperature and your lab's
>temperature to that hot spot reading to get an estimation of the hot
>spot. The issue with this method is that it does not take into account
>the added resistance and hence loss power which will make the actual hot
>spot even higher. Thus it is a linear approximations good only for
>small differences between your lab's temp and the customer's speced
>ambient. You can make a thermal chamber and characterize the
>transformer's performance over load and ambient temp by embedding
>thermocouples in the winding. That would up the cost to the customer
>but if the order is large, an instrumented "gold standard" is a good
>thing to have around anyway! Whose formulas are you using and how
>inaccurate are they?
>
>Tomm
>
>Will Matney wrote:
>
>>Tomm,
>>
>>Yes, it sure does. I have done that in the past, and even snaked a temp
>wire into a coil when winding it then pulled it out later on. The hottest
>spot in each coil is in its center and mid ways of each coil. This being
>for the primary and each scecondary. The coolest coil will be the outer
>one, and the hottest closest to the core. There is a long drawn out
>formula to supposedly determine this where you use the thickness of the
>insulation between layers and between each coil along with the hot
>resistance. I found though that it isn't as exact as they say it is
>because there's just too many variables involved, and one coil can be
>different than another even though the math says it's not so. What I'm
>wanting to do is get a good average of temp say for one model transformer.
>There are some purchasers who want to know the hottest a transformer will
>get including using the temperature of the enviroment where it is to be
>installed. I found that using the temp estimating formula just
>> dont jive with the real world though, and am looking for ways to
>measure this and get a good average. Then new designs can be built using
>this data to where it would be more thruthful in what the actual loaded
>temp is. I'm writing a transformer program to calculate power
>transformers, and the temp part I'm wanting to modify, or not include if I
>cant figure out the fudge factors. Any ideas?
>>
>>Best,
>>
>>Will
>>
>>
>>*********** REPLY SEPARATOR ***********
>>
>>On 1/29/06 at 4:29 AM KD7QAE wrote:
>>
>>
>>
>>>Will,
>>>
>>>It would be much simpler to make a jig to load the xfmr and wait until
>>>the temp on the outside stabilized then switch over to an ohmmeter to
>>>read the hot resistance of the coil in question. Comparing the cold and
>>>hot resistance will give you a good average temp of the coil under
>>>test. knowing the outer temp is good because you can now do a simple
>>>estimation of hot spot by taking the difference between the average and
>>>the outer temp (assume it is the coolest) and add that to the average to
>>>get the hot spot temp. Make sense?
>>>
>>>Tomm
>>>
>>>Tomm
>>>
>>>Will Matney wrote:
>>>
>>>
>>>
>>>>We had discussed using the new infrared aimable thermometers for this.
>>>>
>>>>
>>>You aim the laser dot and read the temp. I was thinking Rich Measures was
>>>going to try this out using a Fluke model. If it was a sucess, I would
>>>like to know the outcome. I myself though of using one to check the coil
>>>temperatures on transformers running under full load. The outter coil
>temp
>>>can be used to extrapolate the internal temp at the mean radius of each
>>>coil this way. Has anyone tried this with any success yet?
>>>
>>>
>>>>Best,
>>>>
>>>>Will
>>>>
>>>>
>>>>*********** REPLY SEPARATOR ***********
>>>>
>>>>On 1/28/06 at 8:00 PM Bill Turner wrote:
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>ORIGINAL MESSAGE:
>>>>>
>>>>>At 11:35 AM 1/28/2006, Gary Smith wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>Top of my one of my 3cx800 a7's reads 73 degree at idle in standby
>>>>>>Temp of tube key down 145 degrees. 1kw in rtty out.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>
>>>>>The top of the tube is not especially critical. It's the seal
>>>>>wherever metal meets glass or ceramic that counts. If you can reach
>>>>>that with the tube operating, fine. Many rigs you can't.
>>>>>
>>>>>73, Bill W6WRT
>>>>>_______________________________________________
>>>>>Amps mailing list
>>>>>Amps@contesting.com
>>>>>http://lists.contesting.com/mailman/listinfo/amps
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>
>>
>>
>>>>_______________________________________________
>>>>Amps mailing list
>>>>Amps@contesting.com
>>>>http://lists.contesting.com/mailman/listinfo/amps
>>>>
>>>>
>>>>
>>>>
>>>>
>>
>>
>>
>>_______________________________________________
>>Amps mailing list
>>Amps@contesting.com
>>http://lists.contesting.com/mailman/listinfo/amps
>>
>>
>>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|