SE 813 OPT..Am I On the right Track?

[valveitude]

First off, thank you Yvesm!! Your OPT Design Assistant program is great. I, like a lot of people, was starting to pull my hair out wading through all the differing design approaches published on the web and elsewhere. Being able to see all the parameters change with subtle changes of the design without blistering your calculator finger is fantastic. There are two other freeware programs that make up my opt software kit, and I recommend them highly. First is Wiretrons wire program available here:

http://www.wiretron.com/free.html

It has a ton of useful wire info and conversion utility. The other is Josh Madison’s Convert program, available here (get the zipped version):

http://www.joshmadison.com/software/convert/download.asp

Its invaluable for converting back and forth from different measuring systems.

I was given an industrial 3-phase motor controller to harvest parts, in this goldmine were 2 chokes with this core…

An externally hosted image should be here but it was not working when we last tested it.

I have no way of knowing if they are m6 or not, but they do appear to be grain oriented. The laminations are .025” (.6mm), and as a bonus, were assembled with an air-gap, so the lams are already set up for an SE opt.

Here is a screenshot of what I think is the best design compromise…

An externally hosted image should be here but it was not working when we last tested it.

Not knowing if it is m6, I am assuming it isn’t, and, took a conservative approach with the Bmax. I read somewhere that a lower Bmax also minimizes the effect of thicker laminations. Is this correct? It seems to make sense.
One of the things I really appreciate with the program is the “Actual Primary Z” window. I was having a hard time trying to fit the windings with a target of a 5k primary, when I realized that lowering the target actually got me closer to what I wanted (5k-6.5k).
The Fo is a lower than I like, but I plan on winding in 2 sections per Radiotron. This will lower the stray capacitance by a factor of 4.

This is my planned winding scheme…

An externally hosted image should be here but it was not working when we last tested it.

Because of the 800-1200 volt B+, I am going to pot the transformer. I plan to use wax, either beeswax or paraffin. Anybody have any recommendations as to which is the better choice, or could suggest something else?

If anybody sees anything I’ve overlooked, or could do better, please let me know. I’ve attached the project file (mox) if anybody wants to have a go. The core data is in the above screenshot

Thanx,
Casey

 

[jlsem]

I have no way of knowing if they are m6 or not, but they do appear to be grain oriented. The laminations are .025” (.6mm), and as a bonus, were assembled with an air-gap, so the lams are already set up for an SE opt.

If they are grain-oriented it should be a fine output transformer. M6 laminations would be thinner but if you Z-wind the bobbin, high-frequency response should be good.

Because of the 800-1200 volt B+, I am going to pot the transformer. I plan to use wax, either beeswax or paraffin. Anybody have any recommendations as to which is the better choice, or could suggest something else?

Microcrystalline wax is the usual and much less expensive than beeswax. It's available here:

http://www.freemansupply.com/MicroCrystallineWa.htm

John

 

[valveitude]

M6 laminations would be thinner but if you Z-wind the bobbin, high-frequency response should be good.

Thanx for the reply John. I'm not clear on the term "Z-wind", looking at my winding diagram above, the primary does resemble a sideways "Z". Is this what you mean?

Microcrystalline wax is the usual and much less expensive than beeswax

Thanx for the tip. I'll check the price and see how it compares to beeswax around here. There are a lot of bee farms around hear (Eastern Washington) and I'm going to see what the bee combs cost getting them direct from the source.

Casey

 

[NickC]

Hi
The laminations looks like a very bad transformer size. 0.6mm i think it is even bigger than normal power trans grade of m19 at 0.5mm only.
I suggest for your 813 you further reduce your bmax to preven core saturation.

Pot helps but beeswax is better but at a very much increase in cost. Paraffin is much cheaper. Also beeswax is often used to prevent corona discharge at that voltage. I suspect even parafin help too. ANother wax to look into is microcrystalline wax. No bad good too. Bartolucci uses them too.

 

[Yvesm]

On Tracks !

Hi Valveitude !

First, apologies for delay (was off home) and thanks for apreciation !

I have no way of knowing if they are m6 or not, but they do appear to be grain oriented. The laminations are .025” (.6mm), and as a bonus, were assembled with an air-gap, so the lams are already set up for an SE opt.

No, there are certainly not M6x (too thick) but, in an SE design, the permeability is already killed by the gap, and the only remaining limitation is the Bmax.
Lo grade iron saturates at lower B, but less abruptly than M6x, and you set lowest freq at 20 !!

Here is a screenshot of what I think is the best design compromise…

Looks fine, just why do you use 8 Ohms in the "reflected load" when you set the secondary at 6 ?

Also check at what happens if you set the tube Rp at some 5000 ohms (unless you use the 813 triode strapped ?)

I beleive that with such size of tranny, the "dual bobin" arrangement is the best to lower the parasitic cap, even if there is room left to increase insulation thickness (or have access to some TEFLON foils).

Sorry for not having (yet) introduced this option in the program !

Anyay, your on the tracks, for sure.

Yves.