Author Topic: Voltage boosting an opamp  (Read 20420 times)

0 Members and 1 Guest are viewing this topic.

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Voltage boosting an opamp
« on: December 29, 2016, 09:25:27 pm »
Hello!

Just got me thinking... For some project I might need a precision DC voltage source controlled via a DAC. The problem is I need 4 quadrant operation - so opamp is the obvious solution. The voltage we are talking about might be say few +5V downto -50V, output current 1mA is more than enough, BW not critical, 1kHz very well sufficient.

The problem is, for +5-50V output I'd need an opamp designed for at least 60V supply operation. I know that some high voltage opamps do exist, but mostly expensive. I'd like to try to stick with jellybean parts for this, not afraid the component count will be a little higher. Only if that fails, I'll buy a high voltage opamp for this :-)

Do you have any suggestions, how to voltage boost the output of an opamp?

One of my first ideas was to ground the output of the opamp through a rather small impedance and then to use a pair of current mirrors in the opamp supply rails, to control the output voltage.
I have drawn my idea, please see below. Does it stand a chance to work?  I think it just might, but I may well be mistaken. 

Thank you for ideas as always!
Yansi

//EDIT: Just found this voltage booster (the second image below). Almost exact schematic! Thats pleasing I have reinvented the wheel and it might work just fine. Also interesting to note the local feedback through R1 C1.
Still I do not get the purpose of matching/pairing the trannies in here. What will the assymetry cause? I can't figure out.
« Last Edit: December 29, 2016, 10:18:30 pm by Yansi »
 

Offline f5r5e5d

  • Frequent Contributor
  • **
  • Posts: 349
Re: Voltage boosting an opamp
« Reply #1 on: December 30, 2016, 06:30:58 am »
http://www.linear.com/product/LTC6090 is a bargain at $5 if you value your time at all
 
The following users thanked this post: Zero999, julian1

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #2 on: December 30, 2016, 01:59:51 pm »
$5 is not bargain. I can make at least 5 of those circuits above  for $5, with the same result. Sorry.

And btw, the availability of LTC6090 at local major distributors is exactly zero.
« Last Edit: December 30, 2016, 02:01:47 pm by Yansi »
 

Offline KhronX

  • Frequent Contributor
  • **
  • Posts: 345
  • Country: fi
    • Khron's Cave - Electronics Blog
Re: Voltage boosting an opamp
« Reply #3 on: December 30, 2016, 02:37:22 pm »
That second schematic makes a lot more sense - i was wondering where the opamp's output was going to (or how it was working at all), in the hand-drawn one :)
Khron's Cave - Electronics - Audio - Teardowns - Mods - Repairs - Projects - Music - Rants - Shenanigans
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #4 on: December 30, 2016, 04:25:26 pm »
After a few (two or three?) hours of breadboarding, I would  prefer using a common base gain stage used for this purpose, instead of this fiddly crazy supply current sensing topology. I couldn't manage it to work good enough if at all.  The common base gain stage works more than good enough, however for AC signals, I'd be very afraid of significant crossover distortion.  :-//
This time I just need to amplify a slowly varying DC signal.

Now just only current limiting to be  added to that circuit.
« Last Edit: December 30, 2016, 04:38:11 pm by Yansi »
 
The following users thanked this post: TiN

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 2077
  • Country: us
    • The Messy Basement
Re: Voltage boosting an opamp
« Reply #5 on: December 30, 2016, 07:38:11 pm »
The supply current configuration originates from some very old app note or manufacturers catalog. It's trouble-prone and Rod Elliot takes a shot at it on his site, suggesting any more standard topology is better. I had some success with it many years ago, probably with a simple 741 and some T05 devices, but also burned up a lot of parts!
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #6 on: December 31, 2016, 12:29:02 am »
Thanks for the tip, I will try to look it up on Rod Eliott's website.

Now that I have successfully built a prototype DC/DC converter for the project, I will try the above topology (common base gain stage) with the full voltage span (requirement for the output voltage is like +5V downto -50V, currently).

The converter is based on LM5007, input voltage is +12V, outputs -56V up to 1W.
« Last Edit: December 31, 2016, 12:37:06 am by Yansi »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Voltage boosting an opamp
« Reply #7 on: December 31, 2016, 12:29:57 am »
That second schematic makes a lot more sense - i was wondering where the opamp's output was going to (or how it was working at all), in the hand-drawn one :)

Either works but the second one uses the operational amplifiers output as a current feedback input to better control the gain of the external transistor stage.  When properly compensated, this is a great circuit.

The supply current configuration originates from some very old app note or manufacturers catalog. It's trouble-prone and Rod Elliot takes a shot at it on his site, suggesting any more standard topology is better. I had some success with it many years ago, probably with a simple 741 and some T05 devices, but also burned up a lot of parts!

I have had very good results with three input and 1 or 2 supply line output operational amplifier configuration.

I would like to see the application note where it originated.  The oldest reference I have seen is in the Tektronix 7A29 vertical amplifier from 1979 where it was used with 741s.
 

Offline f5r5e5d

  • Frequent Contributor
  • **
  • Posts: 349
Re: Voltage boosting an opamp
« Reply #8 on: December 31, 2016, 01:56:37 am »
single ended Class A with a 2 ma ccs at -50 V wouldn't have any crossover problem, use fewer Q

you really need a load spec - too much Cload and compensation becomes more "interesting" - would require some better modeling to recommend the shown feedback lead C and dominant pole integrator



Quote
...$5 is not bargain....

...After a few (two or three?) hours of breadboarding...

Quote
http://www.linear.com/product/LTC6090 is a bargain at $5 if you value your time at all
:)
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #9 on: December 31, 2016, 02:05:47 am »
Electronics is my hobby and 2 hours spent breadboarding something is more valuable, than trying to buy $5 opamp from LT website. Oh, you just forgot there is $250 minimum order limit and something like $50 shipping + VAT and dealing with customs.  Really no  thanks.

I have tried the single ended class A with CCS few times. Didn't work as expected, stability problems. The CCS makes it really high gain stage, the opamp didn't like it. Still, class A with CCS (standard "discrete VAS" ) doesn't solve much, if both voltage polarities are required (how will you interface the opamp output to the class A transistor thats near the V+ or V- rail respectively)

Can you show a schematic, how would you do it?
 

Offline f5r5e5d

  • Frequent Contributor
  • **
  • Posts: 349
Re: Voltage boosting an opamp
« Reply #10 on: December 31, 2016, 02:34:37 am »
spec the load, the supplies you expect to work from and I could sim up something

Quote
  ...The CCS makes it really high gain stage...
not really, the common emitter output from the complementary Q collectors is just as high gain in your earlier circuit too


but I won't build/debug/compensate when there is a $5 solution - I expect to spend to achieve even hobby goals and as a comfortably retired US EE that threshold is a lot more than minimum wage

« Last Edit: December 31, 2016, 03:44:18 am by f5r5e5d »
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7043
  • Country: nl
Re: Voltage boosting an opamp
« Reply #11 on: December 31, 2016, 03:22:15 am »
The common base gain stage works more than good enough, however for AC signals, I'd be very afraid of significant crossover distortion.

Why not just bias the whole thing to have quiescent current then? (Needs some more resistors and the first stage bases can no longer be connected to ground.)
« Last Edit: December 31, 2016, 03:24:42 am by Marco »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Voltage boosting an opamp
« Reply #12 on: December 31, 2016, 03:40:18 am »
I have tried the single ended class A with CCS few times. Didn't work as expected, stability problems. The CCS makes it really high gain stage, the opamp didn't like it. Still, class A with CCS (standard "discrete VAS" ) doesn't solve much, if both voltage polarities are required (how will you interface the opamp output to the class A transistor thats near the V+ or V- rail respectively)

Can you show a schematic, how would you do it?

Take a look as how analog oscilloscope z-axis amplifiers work.  They are discrete and very fast but if the single transistor error amplifier was replaced with an operational amplifier, then they would look like this.  One side of a horizontal amplifier usually works the same way.

They both generally use a constant current load and shunt feedback to make an inverting transimpedance (current to voltage) amplifier.  Tailoring the output voltage range requires a level shift to drive the output transistor but that will not be required if the output transistor is in range of the operational amplifier.
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #13 on: December 31, 2016, 11:08:56 am »
evb149:  I know these supply voltage bootstrapping apnotes, but I don't like the idea do to the possible latchup problems (supply voltage sequencing) and other limitations.

Yes, the schematic was cropped from that one (SNOA600B). Sure, can add other biased stages, but there is a lot of trannies extra, that I think the DC amplifier won't need.  Would rather do what Marco suggests and that is to bias the complementary pair with the common base. I think for the simpliest, a pair of diodes in between the bases, biased with a pair of resistors to V+ and V-. Like this, please see the attachment.

I will response to the rest after a lunch.



 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7043
  • Country: nl
Re: Voltage boosting an opamp
« Reply #14 on: December 31, 2016, 12:18:41 pm »
If you use a string of 4 diodes instead and give all the transistors emitter resistors the biasing is more stable and you get current limiting for free (the transistor stage is actually a voltage to current amplifier, with those resistors the opamp swing limits the current swing).
« Last Edit: December 31, 2016, 12:22:23 pm by Marco »
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #15 on: December 31, 2016, 12:39:57 pm »
Marco: Current limiting of what will I get?  The current through the common base stage is limited by the 10k  resistor. Moving the base up a diode or two won't change much, nor will it add current limit to the output stage. Or have I missed something?


David Hess: The analog scopes I know and I have worked with, were mostly fully discrete and used cascode stages, using a small signal NPN on the bottom, and a high voltage one on top of it. I have never looked through the Z input stages, but have some schematics still printed out here, so will take a look.

Here's what I have come up with. Added current limiting for the output stage. Little biasing for the common base stage to minimize the emitor voltage "wiggle" between the Q1 and Q2 opening.  Also changed the feedback network by adding the 100ohm series resistor on the output - it should make it stable into capacitive load, while maintaining precise DC voltage accross load. (that also is a response for f5r5e5d about the load spec) What do you think about it?


 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7043
  • Country: nl
Re: Voltage boosting an opamp
« Reply #16 on: December 31, 2016, 01:38:31 pm »
Marco: Current limiting of what will I get?

Output current. As I said, the transistor stage is a V to I converter. If you add emitter resistors it's a pretty linear V to I converter in fact, so with a limited voltage swing you get a limited output current swing. In the case of my pic you get about 10 mA per Volt.
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #17 on: December 31, 2016, 01:54:26 pm »
In that case the current limit is NOT because of adding the diode bias.  The output current is limited due to adding emitter resistors for the output transistors - which of course can be done also in my schematic, even in the cropped original one from the application note.
But you haven't said anything about emitter resistors, did you? That's why I haven't understood you.

But thanks for the tip, I can eliminate out the Q5 and Q6 from my last schematic and limit the output current by choosing a suitable resistors in the rest of the schematic.

EDIT: Doh! Silly me! I read wrong. U said give all the transistors emitter resistors... then you were of course right!
« Last Edit: December 31, 2016, 01:58:19 pm by Yansi »
 

Offline Brutte

  • Frequent Contributor
  • **
  • Posts: 614
Re: Voltage boosting an opamp
« Reply #18 on: December 31, 2016, 02:08:51 pm »
The problem is, for +5-50V output I'd need an opamp designed for at least 60V supply operation.
with single power supply.
The voltage sink/source always has only two terminals. So to build 4-quadrant one you either need a +5V:-50V push-pull tied to global GND or you can use two push-pulls not referenced to GND (floating, supplied from separate DC source). The advantage of the latter is that the voltage effectively doubles but the supply must not be GND referenced (your op-amp power supply must not be earth'ed). Then a raw LM224 would do.
If #1 outputs 0V and #2 outputs 30V then you get +30V and when #1 outputs 30V and #2 outputs 0V then you get -30V. Just add a fixed -25V offset between GNDs and instead -30:+30 you get -55V:+5V

The 35V limit only applies to Vcc-Vss
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #19 on: December 31, 2016, 02:34:39 pm »
The problem is, for +5-50V output I'd need an opamp designed for at least 60V supply operation.
with single power supply.

Single or not, +5-55V is the same 60V accross opamp as +-30V or whatever else combination.

Floating load is not an option, I need the output referenced to ground. Floating the power supply is also not an option, because the design as a whole will need at least two more floating supplies for some other circuitry, really not a big fan of making stuff with too much complex power supplies.

I do not understand the rest of your comment, as all circuits posted above will work in 4 quadrants, both polarities of voltage and current.
« Last Edit: December 31, 2016, 02:37:49 pm by Yansi »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Voltage boosting an opamp
« Reply #20 on: December 31, 2016, 02:46:24 pm »
David Hess: The analog scopes I know and I have worked with, were mostly fully discrete and used cascode stages, using a small signal NPN on the bottom, and a high voltage one on top of it. I have never looked through the Z input stages, but have some schematics still printed out here, so will take a look.

The cascodes are added to the current source and output transistor to allow high voltage operation with fast transistors and eliminate the effect of Miller capacitance which would otherwise limit high frequency performance.  They could be added to Q3 and Q4 for the same reason but with such a slow error amplifier, they will not be necessary.

In your circuit, I would add small value emitter resistors to Q1 and Q2 to better control the idle current of the common base input stage whether 2 or 4 diodes are used.
 

Online YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #21 on: December 31, 2016, 03:48:15 pm »
Sure, that is also one of the clever methods of improvement some members suggested me when I was designing the regulated high voltage power supply - ie not to drive the high voltage power mosfet gate directly from an opamp, but to use a cascode with a smaller suitable low voltage BJT.

Yeah, shall add small emitter resistors there to be on the safe side.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Voltage boosting an opamp
« Reply #22 on: December 31, 2016, 05:24:55 pm »
Marco: Current limiting of what will I get?

Output current. As I said, the transistor stage is a V to I converter. If you add emitter resistors it's a pretty linear V to I converter in fact, so with a limited voltage swing you get a limited output current swing. In the case of my pic you get about 10 mA per Volt.
Very good, except it doesn't have any voltage gain!

Increase the values of R4 & R3.


Yes it does. It's just the common base stage doesn't. The preceding common emitter stage does have plenty of gain.

The circuit only needs to be able to provide 1mA so all the resistor values can be higher.

As the output voltage swing is +5V & -50V, the bandwidth is low and linearity isn't important, then the gain could even be asymmetrical, with more gain on the negative side than the positive.
« Last Edit: January 01, 2017, 02:34:52 pm by Hero999 »
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Voltage boosting an opamp
« Reply #23 on: January 01, 2017, 10:57:11 pm »
The power supply doesn't have to be +/-55V. +15V & -15V  will do.

 

Offline f5r5e5d

  • Frequent Contributor
  • **
  • Posts: 349
Re: Voltage boosting an opamp
« Reply #24 on: January 02, 2017, 01:24:18 am »



why complicate the output?

you really only need SE Class A with 2 Q and simple bias:

the Q1 LED biased ccs on the negative rail

the Q2 folded cascode on the positive end, which is just another simply biased from Vsource(s) ccs and a R to the op amp output that "steals" some current

more "interesting" is how to compensate which really, really needs a load Z spec - with the cascaded gains of op amp and any V boost stage there will always be some C_load that will be unstable

my sim compensation is totally eyeball guesstimation and a few repeated sims looking at overshoot, 1p to 1n Cload - with a load spec simplified loop equations more optimum compensation could be found

« Last Edit: January 02, 2017, 02:44:44 am by f5r5e5d »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf