100 to 1 Transformer Results

   I have put together a 100 to 1 transformer using two 10 to 1 transformers.
In the image posted, I show the circuit diagram a photo of the circuit and a photo of the NanoVna
display with either the 50Ω load or a 2Ω load. 100 to 1 is -40db so it starts out well.
  Do you have any thoughts how to improve the response to make it flat?
Span is 50kHz to 150MHz.
                                          
https://www.dropbox.com/s/kx5yjqxlzbigeuf/100%20to%201%20with%2050%CE%A9%20load%20and%202%CE%A9%20loads.jpg?dl=0

                                                     Thanks, Mikek

On 19-10-2022 21:07, Lamont Cranston wrote:
> 
>     I have put together a 100 to 1 transformer using two 10 to 1 transformers.
> In the image posted, I show the circuit diagram a photo of the circuit and a photo of the NanoVna
> display with either the 50Ω load or a 2Ω load. 100 to 1 is -40db so it starts out well.
>    Do you have any thoughts how to improve the response to make it flat?
> Span is 50kHz to 150MHz.
>                                            
> https://www.dropbox.com/s/kx5yjqxlzbigeuf/100%20to%201%20with%2050%CE%A9%20load%20and%202%CE%A9%20loads.jpg?dl=0
> 
>                                                       Thanks, Mikek
It's at low frequencies you have a problem, right?

If so, add more turns

On Wednesday, October 19, 2022 at 6:38:59 PM UTC-5, Klaus Kragelund wrote:

> It's at low frequencies you have a problem, right? 
> 
> If so, add more turns

 The intent of the transformer is to drive  a 1Ω to 6Ω resistance. With the 50Ω load (CH1 input impedance)
The -40db is correct at the start frequency, but then it has a gradual increase of the output voltage all the way to 150MHz. (decrease of loss)
 With the 2Ω load, again it is -40db at the start but this time the output voltage drops and the rises again so at 150MHz it is still down to -50db.
(increase of loss)

 Lamont Cranston wrote:
=====================
> I have put together a 100 to 1 transformer using two 10 to 1 transformers. 

** 10 to 1 turns ratio creates a 100 to 1 impedance change.

So one transformer will do, with a 7:1 ratio.
But you *over size* pot core is unsuitable for such a wide range like 150 MHz.

You can buy such transformers for $5, using miniature ferrite cores with twin holes.

 > https://www.dropbox.com/s/kx5yjqxlzbigeuf/100%20to%201%20with%2050%CE%A9%20load%20and%202%CE%A9%20loads.jpg?dl=0 

** Completley Nuts.

Far too many folk come here asking us how to make their nutty ideas work, instead of the best way to do the job.


....Phil.

On Wednesday, October 19, 2022 at 8:35:11 PM UTC-5, palli...@gmail.com wrote:
> Lamont Cranston wrote: 
> =====================
> > I have put together a 100 to 1 transformer using two 10 to 1 transformers.
> ** 10 to 1 turns ratio creates a 100 to 1 impedance change. 

 I don't want 100 to 1 impedance change, I want 10,000 to 1 impedance change.
This is for a Q meter driver transformer.
   HP does it with a 50 to 1 transformer  (2500 to 1 impedance)
using two cores with different mixes. Range is 22kHz to 70kHz. 
They also use a ~2.5Ω output impedance driver with a  1 milliohm output impedance
from the transformer.
  I tested up to 150MHz, because that is the design goal of the questioner.
For me personally. I don't do anything over 10MHz, and mostly under 2MHz.
But I wanted to see what it did. I might learn something.

> So one transformer will do, with a 7:1 ratio. 
> But your *over size* pot core is unsuitable for such a wide range like 150 MHz. 

 That's interesting, is that to much wire causing high interwinding capacitance?
Or R loss in the wire? (at 2Ω load)


> You can buy such transformers for $5, using miniature ferrite cores with twin holes. 
>
 
Oh, I have a bunch of binocular cores #43 material. Others have used #43 and #2  successfully for
a 50 to 1 ratio up to 30 MHz. Although FT-50 size and FT-68 toroid cores.
Pictures and diagram of build in link.
https://www.dropbox.com/s/madumkmu9uymebh/50%20to%201%20diagram%20and%20finished%20pictures.jpg?dl=0


> ** Completely Nuts. 
> ....Phil.

  Nice to hear from you, thanks for the response.
                                             Mikek

On Thursday, October 20, 2022 at 10:40:02 PM UTC+11, Lamont Cranston wrote:
> On Wednesday, October 19, 2022 at 8:35:11 PM UTC-5, palli...@gmail.com wrote: 
> > Lamont Cranston wrote: 
> > ===================== 
> > > I have put together a 100 to 1 transformer using two 10 to 1 transformers. 
> > ** 10 to 1 turns ratio creates a 100 to 1 impedance change.
> I don't want 100 to 1 impedance change, I want 10,000 to 1 impedance change. 
> This is for a Q meter driver transformer. 
> HP does it with a 50 to 1 transformer (2500 to 1 impedance) 
> using two cores with different mixes. Range is 22kHz to 70kHz. 
> They also use a ~2.5Ω output impedance driver with a 1 milliohm output impedance 
> from the transformer. 
> I tested up to 150MHz, because that is the design goal of the questioner. 
> For me personally. I don't do anything over 10MHz, and mostly under 2MHz. 
> But I wanted to see what it did. I might learn something.
> > So one transformer will do, with a 7:1 ratio.
> > But your *over size* pot core is unsuitable for such a wide range like 150 MHz. 
> 
> That's interesting, is that too much wire causing high interwinding capacitance? 

The main problem is that the ferrite core  is conductive, and current flowing inside the ferrite core is a shorted turn on the transformer.

Manganese-zinc ferrite materials are more conductive than nickel-zinc ferrites, and consequently nickel-zinc ferrites are preferred for applications above about 1MHz.

You can manage inter-windng capacitance to some extent, "banked" winding have less interwinding capacitance than simple layered windings. Transmission line transformers manage the capacitance in a different way, but it woulkd be hard to make one to do what you seem to want to do.

<snip>

-- 
Bill Sloman, Sydney

On 20.10.22 14.39, Lamont Cranston wrote:
> On Wednesday, October 19, 2022 at 8:35:11 PM UTC-5, palli...@gmail.com wrote:
>> Lamont Cranston wrote:
>> =====================
>>> I have put together a 100 to 1 transformer using two 10 to 1 transformers.
>> ** 10 to 1 turns ratio creates a 100 to 1 impedance change.
> 
>   I don't want 100 to 1 impedance change, I want 10,000 to 1 impedance change.
> This is for a Q meter driver transformer.
>     HP does it with a 50 to 1 transformer  (2500 to 1 impedance)
> using two cores with different mixes. Range is 22kHz to 70kHz.
> They also use a ~2.5Ω output impedance driver with a  1 milliohm output impedance
> from the transformer.
>    I tested up to 150MHz, because that is the design goal of the questioner.
> For me personally. I don't do anything over 10MHz, and mostly under 2MHz.
> But I wanted to see what it did. I might learn something.
> 
>> So one transformer will do, with a 7:1 ratio.
>> But your *over size* pot core is unsuitable for such a wide range like 150 MHz.
> 
>   That's interesting, is that to much wire causing high interwinding capacitance?
> Or R loss in the wire? (at 2Ω load)
> 
> 
>> You can buy such transformers for $5, using miniature ferrite cores with twin holes.
>>
>   
> Oh, I have a bunch of binocular cores #43 material. Others have used #43 and #2  successfully for
> a 50 to 1 ratio up to 30 MHz. Although FT-50 size and FT-68 toroid cores.
> Pictures and diagram of build in link.
> https://www.dropbox.com/s/madumkmu9uymebh/50%20to%201%20diagram%20and%20finished%20pictures.jpg?dl=0
> 
> 
>> ** Completely Nuts.
>> ....Phil.
> 
>    Nice to hear from you, thanks for the response.
>                                               Mikek


The 10 kohm impedance is unrealistic at 100 MHz. For any tangible-size
component, a stray capacitance to ground is in the order of 10 pF, which
has about 160 ohm reactance at 100 MHz. You're off by a factor of at
least 100.

There is a good reason why VHF and UHF circuits tend to use 50 ohms.

-- 

-TV