Author Topic: Power driver (or amplifier) for signal generator  (Read 1851 times)

0 Members and 1 Guest are viewing this topic.

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Power driver (or amplifier) for signal generator
« on: August 13, 2020, 06:50:42 pm »
Hi,
I have a Rigol 60MHz signal generator and am looking for a power driver (with or without amplification) with the following specs:

- able to drive inductive loads (with both positive and negative square waves)
- can take inputs of at least 10V
- output current up to 2A preferably 3A
- square wave's rise time (for a 10V square wave) to be ideally less than 50ns, preferably 10ns
- amplification is not needed but would be nice
- ideally (not required) short circuit protected

Questions:
1) what should I look for? Where is the best place to find ready built ones?
2) Any low cost (100-200 USD) part suggestion?

Thank you
« Last Edit: August 13, 2020, 07:14:02 pm by ricko_uk »
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #1 on: August 13, 2020, 07:06:59 pm »
you have a few options:

-Class D amplifiers: basically meant for this task (however I'm not sure about their bandwidth as they're meant for audio applications). They can handle inductive loads (speakers) and have high output currents.

-Class AB amplifiers are your next best bet but require heat sinking for high current loads. If you choose the right transistors, they can definitely reach 60Mhz.

If either of those solutions didn't match your required specs for rise time of square waves, you can use a high speed mosfet driver, which controls a push pull CMOS stage allowing you to drive high currents quickly, just choose fets with low gate capacitance and a fast driver.
 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #2 on: August 13, 2020, 07:11:27 pm »
Thank you OM0222,

cannot use the mosfet drivers because I have positive and negative square pulses (forgot to mention that in the OP - updated now).

Class AB sounds interesting, where can I find low cost ready-built ones?
« Last Edit: August 13, 2020, 07:14:20 pm by ricko_uk »
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #3 on: August 13, 2020, 07:14:50 pm »
you should ideally design your own using BJTs or Mosfets. Class AB also uses two diodes and 2 resistors to bias the drive elements so both conduct around zero volts instead of both being off. I think you can buy off the shelf ones but they're most likely low power options.
 
The following users thanked this post: ricko_uk

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #4 on: August 13, 2020, 07:38:18 pm »
Thank you OM0222
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7458
  • Country: pl
Re: Power driver (or amplifier) for signal generator
« Reply #5 on: August 13, 2020, 09:05:52 pm »
Is it just for square waves and nothing else?
 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #6 on: August 13, 2020, 09:12:46 pm »

Yes, square waves only but bipolar (going positive, then zero then negative then back to zero). Any suggestions?
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #7 on: August 13, 2020, 09:18:30 pm »
in that case you can skip the diodes and resistors and use a class B amplifier, since crossover distortion isn't an issue and you only want ON, OFF operation. you can use LTSpice to mock up a prototype.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: Power driver (or amplifier) for signal generator
« Reply #8 on: August 13, 2020, 11:22:58 pm »
- output current up to 2A preferably 3A
- square wave's rise time (for a 10V square wave) to be ideally less than 50ns, preferably 10ns

10 to 50 nanoseconds requires from 35 to 7 MHz of bandwidth.  That is not impossible; some function generators have class-ab output stages which achieve that level of performance, but they do it into a series terminated 50 ohm load for 100 ohms and 200 milliamps and not 6.6  ohms and 3 amps.  So that leaves three options:

1. A heroic class-ab, or even class-b, current mode feedback amplifier should be feasible.  Check out figure 12 of Linear Technology application note 21 for a lower current example.  The problem will be finding fast high power PNP output transistors which are not as common as they used to be.  50 MHz ring emitter transistors might be sufficient at the lower end of the performance range but the higher end is going to require 200+ MHz transistors.  Several will have to be used in parallel which is just as well because of the power dissipation requirements but the physical layout will be critical.

2. Open loop pulse amplifiers are feasible using a diode bridge to switch current sources, but the 2 to 3 amp requirement means it will draw a lot of power.

3. What I might try is paralleling like 4 x LT1210 integrated current feedback amplifiers in a symmetrical quad arrangement.  If each drives 4 ohm ballast resistor, then that yields a 1 ohm output resistance.
« Last Edit: August 14, 2020, 04:16:47 am by David Hess »
 
The following users thanked this post: thm_w

Offline MasterT

  • Frequent Contributor
  • **
  • !
  • Posts: 851
  • Country: ca
Re: Power driver (or amplifier) for signal generator
« Reply #9 on: August 14, 2020, 01:30:21 am »
If load is not grounded, than H-bridge motor driver will do, it was design to drive amps into coils
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #10 on: August 14, 2020, 10:41:49 am »
just as a sanity check, this is what you're after, correct?

Circuit link

If so you can start choosing the parts that meet your specs.
« Last Edit: August 14, 2020, 01:29:12 pm by OM222O »
 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #11 on: August 17, 2020, 11:58:45 pm »
Thank you OM0222, David and MasterT !! :)

Interesting options!!

OM0222, yes that is exactly what I was looking for. Ideally as ready made low-cost box but then again, it seems a quick and easy solution to implement with components too. That's exactly it!

Thank you all :)
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #12 on: August 18, 2020, 08:06:19 pm »
There'd be major issues with rise time if you want to drive inductive loads. You'd probably need to use higher voltages for faster rise time and that starts to rule out moafets and enter BJT territory.

Do you know the specs of the the coil? Aka the inductance and resistance?
« Last Edit: August 18, 2020, 08:08:07 pm by OM222O »
 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #13 on: August 20, 2020, 04:16:52 pm »
Thank you OM0222,

I just got hold of the coil and measured it. DCR is 26R and inductance is 94mH at 1KHz and 46 at 10KHz (the frequencies that my meter uses).

I was also searching for some systems that apply higher voltages to start with to rise the current fast and then switch to lower voltage once the current has reached a certain level. But could not find any. Also I think to make them work for bipolar might complicate things quite a bit?

Thank you
« Last Edit: August 20, 2020, 04:18:58 pm by ricko_uk »
 

Offline Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1478
  • Country: ua
Re: Power driver (or amplifier) for signal generator
« Reply #14 on: August 20, 2020, 04:25:37 pm »
10 ns, 3 Ampere and 94 mH?!  :wtf:  :popcorn:
Have you made some approximate calculations?
Voltage across an inductor is
V = L di / dt = 0.094 H × 3 A / 10e-9 s = 28200  kV = 28.2 MV
 :popcorn:
It will be hard to realize all those requirements simultaneously.
« Last Edit: August 20, 2020, 04:44:33 pm by Vovk_Z »
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Power driver (or amplifier) for signal generator
« Reply #15 on: August 20, 2020, 05:29:15 pm »
New circuit with BJTs as the drive stage and the correct load. The old driving stage is on the right side so feel free to drag and swap them around.

http://tinyurl.com/yy5j3ocg

26\$\Omega\$ at 3 amps is 78 volts ... not sure why you'd have such a high requirement, especially at those rise times: simply won't work.

do you need all of those requirements or under different loads?

A bipolar system is not more complicated than a unipolar one if you can find the right parts to handle voltages and currents.

For the adjustable voltage idea: it's very doable, use a switching regulator, preferably a buck converter and a digital pot that allows you to change the feedback network and set the voltage but it'd definitely won't be possible at the 10ns window. do you need all of the requirements at once?
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11714
  • Country: my
  • reassessing directives...
Re: Power driver (or amplifier) for signal generator
« Reply #16 on: August 20, 2020, 05:58:39 pm »
New circuit with BJTs as the drive stage and the correct load. The old driving stage is on the right side so feel free to drag and swap them around.
http://tinyurl.com/yy5j3ocg
to my knowledge, that simplistic stage wont cut it if 10-50ns Tr is required. proper biasing need to be implemented to avoid full turning off of the bjt on either side at any time. but thats the easy part, the real deal is finding the bjt in real not sim world that can handle such BW at the mentioned power and load type requirement. 2cnts.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf