[TCML] Sec design trade-offs and considerations

[Finn Hammer]

Chris,

You haven`t built enough momentum for suspense, but so far you have got 
one of us confused.

Is the 1000uF cap merely a storage and smoothing cap? @250V?

Then dump this 250V into a 10nF cap. I`d be surprised if you can get 
breakout from that. It`s 0.3uJ

About your SISG coils.
You *could* have worked around the problems of syncronising to the mains 
frequency by going full DC, with resonant charging of the tank cap. Like 
I did, and got full flexibility from 1-1200+ BPS.

Another note, When you removed the SIDAC`s and replaced them with 
opto`s, you put the individual IGBT in grave danger. I doubt it would work.
The reason you can stack the SISG primary modules lies in the fact that 
the SIDAC`s clamp the voltage across each IGBT until they turn on. After 
turn on, they do so themselves.
Opto`s can`t do that.

Cheers, Finn Hammer


Chris Swinson skrev:
> Finn,
> 
> I documented many many pages on its operation, it actually begun as a 
> modified SISG, and there is many many pages on that also. I have not 
> updated the current pages as yet, but a lot of it is not valid anymore 
> anyway so there is nothing I can post yet.  You could look over my first 
> notes and ideas which are on my website, but these are only ideas and 
> simulations.. to be honest probably not worth looking at, well unless 
> you want to then sure! you could view my SISG updates there, plus many 
> other ideas. Though after that work, I changed the design again to a 
> high Q system....
> 
> My SISG pages were to fire the "spark gap" on the zero crossing of the 
> AC wave, rather than shorting out the supply rails via 200R like terry's 
> does.... But then I went onto opto drives in parallel with the IBGT's 
> instead of series.. though I aborted it in favour for my new design.... 
> Though I did get all the bits to build it as all documented!
> 
> Though once the design is working to some fashion then at least I can 
> post a known working circuit.... If I post the circuit then it opens ups 
> a lot of talk on it (which can be good) but I don't want to get too 
> involved in chats about it at the moment. I already went though all the 
> if's and but's during the last 8 months myself any currently do not want 
> to open it all up again for debate for another 8 months on a circuit 
> which has not been even tested yet, would be like speculation so I would 
> rather keep it hush hush for now until I have some valid data and 
> experiments at least to put to the group.
> 
> As the design gets tweaked it is not even it its final stages yet, it 
> may not even work, so I do not see any point in taking up bandwidth on 
> it at the moment... its got to the "build it and see" part and even 
> though first trial runs looked promising there is still some way to go...
> 
> I can tell you the max voltage on the "Tank" can be 250VDC, this is an 
> limitation of the capacitors used. Even so, 1,000uF at 250VDC, well no 
> need to state how many joules!
> 
> I worked out that at 25VDC input my High Q system *should* equal a 500W 
> SG coil. Though as this design is totally backwards to normal coils, 
> errors and omissions expected!
> 
> 1,000uF charges the 10nF 1uH primary which gives 2mhz.. the 1,000uf only 
> delivers the Joules to the primary circuit, in a way the 1,000uF is 
> equiv to a transformer supply... Though the 1,000uF is also part of a DC 
> resonant circuit at 400hz... There is not much more to it, but to go 
> over all the operation..... well just look at all my posts so far on SG 
> coils.....
> 
> I do however hope to have it finished in a few weeks. so will post the 
> final diagram then working or not along with all the operation notes I 
> have documented over the past few months...
> 
> No doubt once the circuit is working in some respect, others then can 
> chip into the development and make improvements or suggestions... though 
> for now I have enough confusion about things without going over oddball 
> high Q designs also!
> 
> It is probable that it will look like a OLTC type design, but with high 
> Q in mind... If there is enough interest then I will try and keep the 
> list posted with progress as it happens....
> 
> Sorry to keep you in suspense!
> 
> Chris
> 
> 
> 
> 
> ----- Original Message ----- From: "Finn Hammer" <f-h at c.dk>
> To: "Tesla Coil Mailing List" <tesla at pupman.com>
> Sent: Sunday, November 25, 2007 7:11 PM
> Subject: Re: [TCML] Sec design trade-offs and considerations
> 
> 
>> Chris,
>>
>> I have a problem visualizing how the 1000uF cap takes part in the 2mHz 
>> oscillation.
>> Could you post a schematic?.
>> You mentioned high Q and low voltage. What voltage do you have in mind?
>>
>> Cheers, Finn Hammer
>>
> 
> _______________________________________________
> Tesla mailing list
> Tesla at www.pupman.com
> http://www.pupman.com/mailman/listinfo/tesla
> 
>