5
\$\begingroup\$

I have this prototype: enter image description here

Input signal is from a sine-wave function generator and circuit required output is 0-4kHz 0-36Vrms (0-100Vpeak to peak). This works OK. But a new requirement is added - must be able to provide 200mA output current.

Would this type of configuration work? enter image description here

I have two problems - transistor base voltage too high? I can't find a DC-DC module in Digikey that can do +/-100V 200mA?

Any other possible configuration? I can change opamp if needed.

\$\endgroup\$
4
  • \$\begingroup\$ "...circuit required output is 0-4kHz 0-36Vrms (0-100Vpeak to peak)." That voltage requirement is ambiguous - is the requirement 0-36V RMS? Or 100Vpp? And are you expecting a single-ended output (only positive voltage)? \$\endgroup\$
    – MOSFET
    Commented Nov 14 at 22:57
  • \$\begingroup\$ not sure about single ended definition but max sinewave output need to reach +50V peak and -50V peak. In rms, this is 36Vrms (+18Vrms, -18Vrms) \$\endgroup\$ Commented Nov 14 at 23:11
  • \$\begingroup\$ Power dissipation in transistors Q1 and Q2 is potentially going to be several watts, well beyond what the MMBT39xx can handle. Use something designed for that, which can handle 120V or more, and definitely use a heat sink. Also, see R1 in Spehro's answer which solves a couple of problems with your design. He also uses the surprisingly good and reasonably priced LTC6090 140V op-amp. So your design, with tweaks, is good. \$\endgroup\$ Commented Nov 15 at 5:40
  • 1
    \$\begingroup\$ @ArielNarboada - Hi, Where did this image come from? To comply with the site rule on referencing, details of the original source of copied / adapted material must be provided by you, next to each copied / adapted item. If the original source is online & public, please edit the question & add the webpage/PDF/etc. name & its link (URL) (e.g. website name + webpage title + its URL). If the source is offline (e.g. printed book / private intranet) then edit the question & add full source details e.g. title, author, page, publisher etc. \$\endgroup\$
    – SamGibson
    Commented Nov 15 at 11:41

6 Answers 6

7
\$\begingroup\$

Here is a simple way to do it:

enter image description here

Power dissipation of the transistors is about 2W each with +/-70V supplies and +/-200mA peak output current.

\$\endgroup\$
3
  • 3
    \$\begingroup\$ You beat me to it, same design down to the LTC6090 and R1=200! It's worthwhile noting that R1 is important, but often omitted. It keeps feedback low impedance and negative in the case where both Q1 and Q2 are off, and also really helps with crossover distortion. Still, I was surprised how little crossover distortion at 4kHz this design produces, the LTC6090 is remarkably good, if the simulation is to be believed. \$\endgroup\$ Commented Nov 15 at 5:08
  • 1
    \$\begingroup\$ @SimonFitch Ha. There's a bit of a trade-off with the 20p capacitor. It could probably be slightly lower, but I think there needs to be one there. The LTC6090 is not insanely expensive either, for what it is. \$\endgroup\$ Commented Nov 15 at 5:13
  • 1
    \$\begingroup\$ I did not include C1, the step response wasn't excessively wobbly without it. 20pF produces very nice damping though, I think it's spot on. \$\endgroup\$ Commented Nov 15 at 5:26
4
\$\begingroup\$

Would this type of configuration work?

There is going to be some cross-over distortion when the op-amp output has to jump between ~ +0.7 V and -0.7 V at each zero crossing --- since the op-amp can't do that instantaneously there will be a "dead zone" in the output waveform at each crossing.

If your application would be impacted by this distortion, you can reduce it with a simple adjustment:

schematic

simulate this circuit – Schematic created using CircuitLab

This does cost some power as R1 and R2 must be chosen low enough that they can provide sufficient bias current to the BJT bases and keep the diodes forward biased.

\$\endgroup\$
3
\$\begingroup\$

If your requirement is 36 \$V_{RMS}\$ you don't need a +/- 100 V supply. That's 100 \$V_{pk-pk}\$ which is +/- 50 V so you just need that plus a bit more for some headroom. You might want to go with the same voltage as the op-amp is running from so you don't get the base-collector junctions reverse biased.

As for the buffer, you would need not only higher voltage transistors but higher current and wattage ones. 50 V @ 200 mA is 10 W so the transistors are going to need to be a bit beefier than the ones shown. And you need to make sure the op-amp can drive them with enough base current or else use Darlingtons.

\$\endgroup\$
2
  • \$\begingroup\$ i can't find a suitable power supply. only this which is 80mA max: digikey.com.au/en/products/detail/analog-technologies-inc/… \$\endgroup\$ Commented Nov 14 at 23:13
  • \$\begingroup\$ @ArielNarboada, there are 100's of 1000's of power supplies out there on the market. You can find one for this project if you look hard enough. You could just find a transformer with a different winding ratio in your existing design and change the output voltage that way. \$\endgroup\$
    – The Photon
    Commented Nov 15 at 2:21
3
\$\begingroup\$

The output voltage and power requirement probably means you need a more robust output stage than can be achieved with 2N3904/3906 alone. A second driver stage can help (simulate it here):

enter image description here

Design Notes:

  • This is basically an audio amplifier with a gain of 50. It's shown with a 50 ohm load which you would expect a function generator to drive.
  • Op-amp shown in the sim is a 741, but can be any decent audio-grade type. It only needs to swing a couple of volts.
  • The 1k/100 ohm voltage divider at the op-amp out makes the swing larger and reduces crossover distortion at the amp. The diodes also reduce crossover distortion at the driver.
  • The first stage transistors can be 2N3904/3906. The second stage should be something bigger, able to handle 6-7W or so.
\$\endgroup\$
2
\$\begingroup\$

The LTC6090 datasheet](https://www.analog.com/media/en/technical-documentation/data-sheets/6090fe.pdf) gives several applications of "bolting" transistors to the device to get a higher current output. For instance, there's this high current pulse amplifier: -

enter image description here

Or, there's this 100 watt audio amplifier: -

enter image description here

The datasheet even shows you how to current limit the output: -

enter image description here

\$\endgroup\$
2
\$\begingroup\$

My reaction when reading the question:

I can't find a DC-DC module in Digikey that can do +/-100V 200mA?

What you want is an amplifier. DC/DC converters are typically power converters, and do not work as amplifiers. Any amplifier that can work with the swing and current you need will work fine. 36V, 200mA, means about 7.5 Watts, so go for at least 10 Watts rated capacity.

In addition to choosing an amplifier with the appropriate rated power and low distortion, beware of output filtering requirements. Most class D amplifiers need heavy filtering, which is implied to be provided by a speaker/driver on the output, but if you're just outputting signal into a resistive load, you will need to add additional filtering yourself in that case.

Main challenge would be to find an amplifier with +/- 50V range. You might need to get something like an LM3886 and bridge it to get where you need to go. Also, limiting to 200 mA current at 36 Volts, you'd need to specify minimum output impedance of 180 Ohms.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.