Driving a 3D coil with the AD9833

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
Oh right. I removed the "smoothing capacitor" and increased C2 to 100nF. Works great in the simulation.

The signal is 2V peak to peak measured with the capacitors for the resonance. Which I think is right to take into account, right ?
 
Last edited:

Ian0

Joined Aug 7, 2020
1,922
Oh right. I removed the "smoothing capacitor" and increased C2 to 100nF. Works great in the simulation.

The signal is 2V peak measured with the capacitors for the resonance. Which I think is right to take into account, right ?
Fair enough, if the receiver is that good, you might have to attenuate it! Can the SA612 cope with 2V p/p?
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
Alright so I got everything working quite good. The current problem I have is that I want to generate the oscillating signal with a chip (I tested the circuit with a signal generator until now). You suggested a counter a few pages back. I looked a bit more and found the si5351. This breakout board is easy to use and has libraries for it. Furhtermore it can generate 3 independant clocks over a wide range. So I programmed it to output 20kHz from 0V-3.3V square. But since the SA612 needs a 250mVpp I need to attenuate the signal. I tried a voltage divider, but it doesnt work really well. I fed the 3.3V square through 25k ohms in series with 2k ohms. But the signal is practically gone. What am I doing wrong ?

-8
 

Ian0

Joined Aug 7, 2020
1,922
Without studying the data, I would guess that it is two things:
1) You have not allowed for the input impedance of the SA612, which is quite low and appears in parallel with your 2k resistor
2) The SA612 sets its own DC bias, so inputs have to be AC coupled. i.e. you need a capacitor between your attenuated signal and the SA612 input, value: not critical, but so its impedance at the operating frequency is negligible. If you DC couple the input, you are forcing the bias point to be somewhere where it doesn't work.
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
For the second point: I forgot to add the schematic. I attached it now. I do AC couple the input with a 10nF.

And then for the first point. Does that mean I need to pick smaller resistors ?
 

Attachments

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
Ok I don't understand. I attached a picture so I can follow a bit better. The way I understand you is that I should remove R2 and change R1 so that R1 and R3 (input impedance, what you described as Rin ?) build a voltage divider.
If that's right so far, I don't know how I could know Rin.
Anyways I tried a few resistor values and observed the voltage after R1, without any success though.

It would be nice if you could explain that again please.
 

Attachments

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
Alright, so I tested a few configurations without the R2 and measured the voltage between R1 and R3. I calculated R3 to be around 18k ohms. Based on that I chose R1 to be 220k ohms. I measured again between R1 and R3 and it was indeed around 250mVpp.

But now I get the 250mVpp square wave with a DC offset. Why is that and how can I take that out ? Because I think the oscillator signal shouldnt have any offset.
 

Ian0

Joined Aug 7, 2020
1,922
Have a look at the internal circuit of the SA612. Don't you just get the idea that the input on pin 6 looks as though it is possibly about two thirds of the way up?
The output voltage will be above it, and the input section will be below it.
So if you have a DC offset of about 3V to 4V on a 5V supply, it could possibly be correct?
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
Yes that is true. I do get about 4.7V between the capacitor and pin 6. I was unclear about where I measured, I ofcourse have to include the capacitor into my explanation.

I measured between R1 and the capacitor.
 

Attachments

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
If there is a better way of generating a frequency at that amplitude and still have the flexibility to change the frequency. That would be fine too.
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
I swapped it with th si5351 because I could produce 3 independant frequencies with this chip. Does that make a huge difference ?
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
197
Ok, so I did some measurements and found something rather strange. So I don't think there is something wrong with the si5351. I found a datasheet online (attached). On the top of that datasheet is the sa612 with the si5351. I configured mine to be the same, but it still doesn't work correct.
One thing I noticed is that when I setup my function generator to output the exact same as the si5351 would do, it works.
Maybe I should clarify a bit more what I mean by "something is wrong".

When I read the analog signal from my receiver and I'm generating the oscillator signal with my function generator I read a constant signal when I hold the transmitter steady. If I move the transmitter the signal gets smaller/larger depending on the rotation. So that works.
Now, when I generate the oscillator with the si5351 and I read the signal I get a sine wave which is approximately 5 seconds in peroid overlayed with the smaller/larger signal I would normally get. So when I say there is something wrong I mean that the reading is sinusoidal although I hold the transmitter rather steady.

Could that be a problem that's induced by the power ?

EDIT: Solved in this thread.
 

Attachments

Last edited:
Top