Small signal transformer

A

AnneB

New Member
Hi,

I have an existing circuit producing a sine output variable between 100 kHz and 300 kHz, with 100 ohm output impedance and around 10 volts peak-to-peak (+/- 5v).

I really need to step this up by about 10 to around 100 volts peak-to-peak. But I don't really think I need a voltage amplifier because the output requires essentially no current. It is simply being fed to a pair of air-spaced plates to generate an electric field. Given the area/spacing of the plates, it works out to around 6pF of capacitance. So at 300kHz that is around 80 Kohms impedance.

Ideally, I would just like to use a little high-frequency signal transformer for this. Does anyone know is such a component available? I have searched for many "pulse" transformers etc etc but struggling to find something with around a 1:10 turns ratio!!

Thanks so much!

Anne
 
Wow!! Thanks for the quick reply Nigel! I have a variety of ferrite cores in my "bits box" -- do you have any guidance on the best core size, number of turns, etc. I am quite happy to have a go at winding something!
 
Are the cores Toroidal? This is what you need preferably, they are also the most efficient means of transformation so the turns/volt is not high at all,
Experiment with a couple of windings, Pri & sec and find the the turns ratio (turns/volt)
 
Thanks Max. I have several small toroidal cores. I will have a go. I am not sure of the correct inductance I would need for the primary so I guess it is a matter of trying it. I was hoping there might be a neat little formula for calculating the turns given the characteristics of the core?
 
Using LTspice, 5Vpk waveform at 100khz. R1 is the 100 ohm impedance of the source. (why is it not 50 ohms?)
L1, L2 is a transformer. C1 is the 6pF load. Teh 330uH is the primary and the 33mH is the secondary stacked on top.
CoilCraft Here is a link to a transformer that is 202uH on the primary. The 5 secondaries need to be wired in series to get a 5:1 boost. It is not 10:1 but is an example of how it could be done.
The coilcraft part has a VT of 35VuS which is about as small as you can go.
 

Attachments

You can make it with high Q or low Q depending on your source impedance and desire to regulate voltage and current when it arcs. The arc impedance drops with rising current.
Big old car distributor coils have a gain of 100 from 50 Ohms with 20 kHz BW. Modern ones have much higher BW.

Impedance ratios will tell me the Q if you want it resonant. If not that's OK too. The Yellow shows 50 Ohms x100 = 5k then if you made it resonant you get another sqrt(impedance ratio) voltage gain at the LC intersection.
 
Last edited:
Thanks Tony
I don't want a resonant circuit as I want to be able to adjust the frequency a bit in my experiments. But the lovely helpful people at CoilCraft have been kind enough to send me samples of their transformers Ron recommended. I will have a play with those and see what can be done.
Thanks again!
 
Anne, you won't be able to avoid the effects of resonance with LC in this f range.

Please try to understand the graph and ask questions.
You must know the fundamentals of impedance Zo and fo and transformer impedance ratios to understand the basics. Then there are eddy currents, hysteresis and leakage inductance effects.
 
ronsimpson said:
What is the inductance of the transformer? on which pins? It looks like the self resonant frequency is below the signal frequency.
Click to expand...
No SRF is above excitation. The secondary AM is the difference frequency.

Shown was the primary left pair as defined (property) = 330 uH and jumpered like an autotransformer but with a slider in the simulation to choose Lp= 0.1u to 330u secondary on the right. Then I appended a 330V arc gap for kicks that otherwise has 0 pF but specs for conduction and leakage R.
 
SRF = 325 kHz , Lp = 330 uH N= 0.1 or 1:10
 
Hi everyone.

The lovely people at CoilCraft delivered me a box of free samples of their transformers! Once I got the secondaries wired properly in series, I have got some great results. Resonance has not been an issue at all. The devices seem pretty low Q so the broadband response is good. I can tune over my entire frequency band of interest with very little loss of gain. So thanks again to Ron for putting me on to these devices.

Job done

Anne
 
The soldering was pretty tricky! It is basically a surface mount device - here is a picture under my microscope!!
 

Attachments

  • Pic01.jpg
    938.5 KB · Views: 130
Sounds like a success story!

Thank you for letting us know how it worked out.

JimB
 
You must log in or register to reply here.

Similar threads

Output transformer of homebrew inverter.
Replies
6
Views
2K
shokjok
S
H
  • Question Question
I need to trick a fluid sensor to return a full signal
2
Replies
24
Views
5K
hhoarder
H
F
  • Question Question
SMPS transformer has square spool
Replies
1
Views
2K
ZipZapOuch
D
  • Question Question
LCD Screen flashing
Replies
5
Views
2K
desferg
D
F
  • Question Question
Transformer diagram incomplete (EFD15/8/5)
Replies
3
Views
3K
OldTechie
O
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…