Hi all,

I have to amplify the signal coming out of ATMEGA32 chip which is a sine wave with peak voltage of 5v and a current in mA. I have developed a circuit using OPAMPS (see attached). I need to amplify the signal to have a peak to peak of 20v and an output current of 100mA. This is used to drive a coil which is represented by the inductor in the schematic attache... I have acheived the voltage amplification (Check Vin_transmitter_coil node) required. But the current i get is only a max of 30mA.. how can i get more current? any ideas?

Also when i connect the inductor load..the Vin_transmitter_coil signal is not a clean sine wave...Is there a way to fix this problem?

 

You forgot to look at the datasheet of the opamp. Like most opamps its output current is a max of only 20mA.

Use complimentary emitter-follower transistors to boost the output current.

 

Started to suggest an LT1010 power buffer, but it just won't quite make it; max limit is 22v.

Here's another way:

Note that peak power dissipation in the transistors is 2.2W. Average is much lower, but not sure just how to do that in LTSpice yet.

I bumped your R1/R2 up; no sense in wasting that much power on feedback.
Note that the diodes must be thermally coupled to the transistors, or things will go south quickly.

 

Hi SgtWookie,

Ur idea is great ..but the output signal gets clipped at 10v..it doesnt reach 10v. What 2 do about tht?

 

You make Vcc/Vee = ± 15V so the clipping occurs at ±13 V

 

But i can only use +/-12v though...

 

So you need a high current rail to rail solution. Is that correct?

 

Hi SgtWookie,

Ur idea is great ..but the output signal gets clipped at 10v..it doesnt reach 10v. What 2 do about tht?

Were you looking at the pink trace? That's the power dissipated in Q1. Inverted, but still the power in Q1 (I got the subtraction reversed)

Oops - after looking again, Q1's power dissipation isn't as bad as I thought it was.
Ie(q1)*(v(n006)-v(Vcc))
Peak power dissipation is 350mW; well within the limits of a 2N2222 or 2N2907.

[eta]
Yes, there is some clipping at 9.5v.

The easiest fix is to increase V2 & V3 to about 13v.

Or you could also reduce R5/R6 to 750 Ohms, which will increase Ib for Q1/Q2. Watch power dissipating in R5/R6/D1/D2 though.

 

Hi again,,

Ya reducing the resistors to 750ohm works good. One other issue is that the frequency of the input signal V1 can change and this circuit needs to give a stable output for this. With the current circuit, the output remains ok for frequencies between 10kHz and 100kHz but for any frequencies outside this range, the output has distortion. What could i do about this?

 

What range of frequencies do you need?

Bandwidth is always limited by something.

 

I need a range of 10KHz to 10MHz. Is this possible?

 

I need a range of 10KHz to 10MHz. Is this possible?

Using just +12v and -12v supplies?

 

Ya,,,,,,,,,

 

Does one end of the coil have to be grounded?
Or can a push-pull driver be used?

 

No coil has to be grounded....

Also, I just realised that the input V1 i gave is a sine wave but the ATMEGA 32 output is a square wave...Is there an easy way to convert a square wave to a sine wave?

 

Not particularly an easy way, but you could construct an LC filter that blocked all of the odd harmonics higher than the fundamental frequency. Of course, you would need to adjust the filter for changes in frequency.

Or, you could use a DAC output to generate a pseudo-sine wave. It wouldn't be too smooth, but it would be an approximation of a sine wave.

What is it that you're trying to accomplish? What is this coil, anyway?

 

Hi,

The coil is a transmit coil of a metal detector model. There is a receive coil that receives the signal and after analysis of the received signal, it is decided whether there is an object or not....

What is the DAC output ?

 

Hi,
I added in an active LPF with a cutoff frequency at around 160KHz to covert the square wave coming from the micro to a sine wave and then passed it on to my previous circuit.....It seems to be working but the output (across the inductor) is not a perfect sine wave a before.

Any suggestions what if anything can be done to improve it?
Thanks

 

Here is the modified schematic

 

Exactly what is your method of metal detection? If you are looking at an eddy current decay time of the receive coil, ie: the TC of the decay being representative of metal being in the loop, the idea is to shut off the transmit coil abruptly and then look at the decay time of the receive coil. Maybe a fast schottky diode across your transmit coil will accomplish this. I also feel that you must use high speed switching transistors to augment the performance.

Keep us informed, DPW [ Spent years making heaters out of op-amps.]