[TCML] quench times again

[Barton B. Anderson]

Hi Chris,

I confused myself, there "is" a relationship with frequency and the 
transfer rate. I'll explain later.

Chris Swinson wrote:
> Ok... So the coil was half the inductance, which halved the mutual 
> inductance which doubled the trasnfer rate which as a result frequency 
> was increases....
If the secondary is wound with very large wire, it's physical size will 
really lower the secondary inductance due to the very few turns allowed. 
Changing the wire size changes the geometry even if the h/d aspect ratio 
remains the same. When I went from 24awg to 18awg in my example, I 
changed the geometry of the wire a little which caused less turns and 
reduced Ls. The fact that Ls changed was the major factor for the sight 
change in the transfer rate.

The relationship with coupling and mutual inductance is 
M12=k*sqrt(L1*L2). So if we change either inductance, mutual inductance 
will change and we would then need to adjust k to get back to the 
original mutual inductance. Javatc equates the mutual inductance and 
then uses K=M12/sqrt.

>
>>
>> To quickly verify, just run my default coil and look at the transfer 
>> time and number of 1/2 cycles. Then, change the secondary wire size 
>> to 18 awg at 516 turns (this will keep the geometry the same and 
>> raise the frequency). You'll find the transfer time will change from 
>> 17.16us to 17.29us (hardly a change). Look how much L changes and yet 
>> transfer time barely changes. You will find secQ increases and ACR 
>> decreases (that big 18 awg wire wound in that geometry is why).
>>
>> Then if you want, use the 24 awg wire and reduce the turns to hit the 
>> same frequency as the 18 awg wire size case above (use 291 turns with 
>> the 24 awg size and secondary change "height2" to 29.55). This will 
>> be a geometric change at near the same frequency. Transfer time will 
>> now decrease to 10.65us. Note what I did here was change the geometry 
>> to cause a transfer time reduction and simply set it for near the 
>> same frequency as the 18 awg frequency using 24 awg wire. In order to 
>> do that, the secondary geometry needed to change (less wire, get 
>> smaller).
>>
> In my testing I kept the coil geometry the same, by using thicker wire 
> and less turns, but physically the coil was the same size...
>
> Repeating my self a little here.....
>
> 200khz  DCR 0.010  L=61uH  ML=134uH  27.5uS
> 400khz  DCR 0.005  L=15uH  ML=33uH   13.7uS
Ok, the large wire size has done two things which affect the transfer rate.
1) L is greatly reduced and this will change k affecting the transfer rate.
2) Frequency is increased and this will decrease the transfer rate.

Here is where I was wrong. Yes, frequency affects the transfer rate. As 
the number of cycles increases, the transfer rate will increase. Here is 
the relationship.

Total Energy Transfer = (0.5/((1/(1-k)^.5)-(1/(1+k)^.5)))*(1/fr)

>
> So 2 things are altering coupling (K factor) then ? one the physcial 
> aspects of the coils, but also the mutual inducation. such as you 
> could  have a 8" dia secondary and 12" primary and just assume this 
> has a K=0.1  , Though are you saying if the secodnary is kept the same 
> size, but reduced in turns and thicker wire, that the mutual 
> inductance would be lower and K would increase ?
Lower k will increase transfer time and lower frequency will increase 
transfer rate. You can affect transfer rate either way. Of course, in 
reality, changing one aspect is going to change more than just the one 
aspect. An example is your own coil with the large wire. Due to the 
larger wire, you have fewer turns, you have big changes in L and C and 
so frequency follows. Frequency is certainly going to affect the 
transfer time as is the coupling which has changed due to the changes in L.
>
>> I just wanted to clarify this situation between geometry (coupling) 
>> and transfer time. Frequency is simply a result of the LC changes. 
>> Note that coupling has increased! Geometry and position is why, not 
>> the frequency.
>
> Right, so I think frequency is related, but its the actual mutual 
> inductance which is the casue of the decrease in mutual coupling....
It's the change in the secondary inductance which changed the mutual 
inductance.
It's the mutual inductance which changed the coupling.
It's the change in the secondary inductance which changed the frequency.
It's the frequency and coupling which changed the transfer rate.

It's all intertwined.
>
> It is confusing..... as you can also reduce coupling by physically 
> making the primary larger which decreses coupling again, but makes the 
> transfer time longer ?!
Yes. Reduce k and transfer time is longer. Increase k, and it's shorter. 
This can be done without changing frequency and usually is.
>
> But if you decrease the secondary turns and use larger wire, it also 
> decreases coupling but energy takes less time
> *confuzzled look*
>
Yes, it does. Inductance is the reason.
The larger wire decreases M, decreases k, and decreases the transfer rate.
Also,
The larger wire decreases L, increases Fr, and increases the transfer rate.

So, both will affect the transfer rate and depending on the "degree" of 
the change, one will affect it more than the other.
>
> I think I see your point to a point... higher frequency really is just 
> a by-product of altering all other factors... all those other factors 
> such as mutual inductance are the real "engine" behind the transfer 
> rate change...
No, I was wrong. I got caught up in terminology of coupling which is a 
product of geometry and inductances and ignored the frequency. In most 
cases we don't make a large enough change to greatly affect the transfer 
rate as much as we do adjusting the coupling. In the case of your large 
wire coil, there is ample change to make a difference and frequency does 
affect the transfer rate quite a bit.

I'm sorry to confuse you on that one. This morning I had to take a 
second look. Glad I did.

Take care,
Bart