I have a cheap signal generator, that can do square and saw-tooth waves up to few MHz. Unfortunately, it can only supply a whisper of power, maybe just few mW.
I would like to use the signal's voltage shape from the signal gen. and supply lot more power (current) to various devices, like solenoids or DC motors etc.
What kind of circuit can take a weak signal from signal gen. and transform it into the exactly same waveform, but with 100 times more current?
That would be complementary transistors, like an output stage of an audio amplifier.
Signal generators are designed to apply small inputs to amplifier stages of receivers etc.
They are not designed for solenoids or motors. Applying a different frequency of a saw tooth would not change the speed of a motor as it has the same power at low or high speed. Same with square waves.
If you are trying to achieve PWM type control, it will not work as the square wave will always supply 50% power.
For the sake of my point I am ignoring frequency response of components when trying to feed motors or solenoids at a couple of MHz!
Weedpharma
A broadband power amplifier module? Do you need the full bandwidth of MHz? - that will
increase the complexity and cost.
Do you care about DC offset? If not then an audio amp will do upto 20kHz.
Yeah, I would not mind going up to MHz range, but kHz would be OK for starters.
In a long run, I would need MHz range. The most important thing for me is to pass through a shape of the waveform from signal gen. without too much distortion.
Its just a question of pointing me into a right direction?
Would I need a different amplifier for each frequency range?
A power op amp may work (quite costly though).
Start with a Kijiji audio power amp. That will give you two channels of maybe 40W each and usable bandwidth to 15kHz, It might have output fuses and everything.
Anything going into the MHz range is a radio transmitter and would be expensive and regulated.
I did not find a good circuit.
But this show what I was thinking : http://www.analog.com/library/analogdialogue/archives/48-03/resolver.html
Scroll down for Figure 11.
Two transistors as output and the output feed back into a OpAmp. Everything is DC.
For some real current, perhaps darlington transistors can be used. The circuit for the base circuit might need some tuning. The more distortion is allowed, the more stable the circuit is.
Just for general sharing, for people who might be doing a similar thing, I found this IC, that can go up to 3A:
You haven't told us what you need. What voltage? What current?
A sawtooth or square wave at a given frequency requires a -much- greater bandwidth than a sine wave.
Gain?
The greater the gain, the lower the bandwidth. GBW aka Gain Bandwidth Product tells you this. For that LM675, the GBW is 5.5MHz. So as a buffer with a gain of 1, bandwidth is 5.5MHz. But with a gain of 10, bandwidth is 550kHz.
But a square or sawtooth also requires a high slew rate, not just bandwidth. A triangle wave or sine wave does not require as high a slew rate. But it depends on the amplitude.
There are ways to combine two Op Amps to get the best of both worlds.
But first we need to know what output voltage, current, frequency, and waveshape, and what kind of voltage the sig gen is putting out.
You need Gain?
Most signal gens. can do 5-10V. So I would like to simply repeat that, so gain, on voltage would be 1.
But on current, I would say would like to go from 20mA on a signal gen. to 2-3A on amplifier. Gain 100-150.
But lets say sawtooth is really important. So current gain 100-150 without too much distortion.
Would that be feasible?
Solenoids will only work at frequencies of a few Hertz or tens of Hertz maximum, so why would you need to feed them with a 1MHz signal?
With an Op Amp, you calculate voltage gain. And you select an Op Amp that will output enough current. You don't calculate current gain.
The LM675 has a power bandwidth of 70kHz, so that's all it'll do.
You really need to quantify the bandwidth you actually need. As JohnLincoln says, a solenoid isn't going to react to a very high frequency. The inductance is too high.
A sawtooth at what frequency? Because you'll need an Op Amp with a very high slew rate and GBW, and power bandwidth to do a sawtooth and a square wave.
gardner:
Anything going into the MHz range is a radio transmitter and would be expensive and regulated.
Any amplifier that can do both high current and somewhat high frequency is regulated? How? Help me out, I've never heard of that.
These parts don't seem to be regulated to buy: http://www.mouser.com/ProductDetail/MACOM/MRF455/?qs=sGAEpiMZZMvDjfggS9kWsfMiuY8Ox1OWZAAEDLIRPWw%3D
Or do you mean regulated as in if you cause interference then the FCC can shut you down? This is a fact for all power levels - evidence of interference is enough. You seem to be saying the part is regulated itself, which I have never seen evidence of. This is the country where you can still mail order 100 round magazines for your semi-automatic rifle and mail order Tannerite. I'd be surprised if you were denied an op-amp because it was both a power amp and could reach into radio frequencies. Or perhaps this is regulation in Europe?
polymorph:
I like that image, it explains so much.
So are there any op-amps better for this task than LM675? That can take more current ( 3A < ) and have a better slew rate?
@Peter_n Darlington transistors are good idea. They can really take a current to the level that suit me and voltage is about right.
What frequency limit?
Or a better question: What kind of rise time at 3A?
3rd edition Op Amps for Everyone
http://www.siongboon.com/projects/2008-04-27_analog_electronics/op%20amps%20for%20everyone%20third%20edition%202009%20(Texas%20Instrument).pdf
4th edition in print, Op Amps for Everyone
Look at page 27, "Power Booster":
http://www.ti.com/ww/en/bobpease/assets/AN-31.pdf
Here is one, 1A output 35MHz bandwidth at gain of 2, and a slew rate of 900V/us.
See page 58, figure 85 for the LT1210 used with two transistors to make a 1MHz bandwidth +-10A output.
http://cds.linear.com/docs/en/application-note/an87f.pdf
Less than $10 each direct from Linear Technology.
