Author Topic: variable gain current amp  (Read 4264 times)

0 Members and 1 Guest are viewing this topic.

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
variable gain current amp
« on: May 14, 2020, 02:48:55 pm »
I am looking to use the https://www.ti.com/lit/ds/symlink/lmp8645.pdf so that i can switch the gain resistor to change ranges. But just to have a gain of 50 I have calculated that i need 250k, to do my more sensitive range i need 25M! have i got it wrong or is this chip pretty hopeless?
 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 11341
  • Country: us
  • $
Re: variable gain current amp
« Reply #1 on: May 14, 2020, 03:28:10 pm »
well its common with modern parts to use high Z there to save power. It's something you would use guard traces with if you follow the datasheets for other parts with similar nuances (i.e. LT)

With the voltage regulators, I think it made sense to use a through hole part with a guard trace under neath IIRC

But those parts were ground referenced. This is floating. It's pretty high (usually the VGA IA's have like 1 meg max)

In the data sheet, maximum gain is 100. Don't know if this corresponds to a noise figure, or just the GBW setting.

You are saying you want a gain of 5000, from a part that has a datasheet gain of 100 max, with 25meg, assuming the formula rgain (25Meg)/ 5k (internal)

Single stage 5000 gain is going to be ugly. It looks like they want maximum to have a value there of 5Meg.
« Last Edit: May 14, 2020, 03:36:03 pm by coppercone2 »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #2 on: May 14, 2020, 03:35:57 pm »
Yea I think basically I need too much gain I'm at 50 with 250k but the other range is 100x higher. Maybe i need to follow with an opamp to do tho 100x
 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 11341
  • Country: us
  • $
Re: variable gain current amp
« Reply #3 on: May 14, 2020, 03:37:04 pm »
Yea I think basically I need too much gain I'm at 50 with 250k but the other range is 100x higher. Maybe i need to follow with an opamp to do tho 100x

you might also get results if you filter prior to heavy gain because its a semiconductor facing something with RF noise that might do weird things, so you go from DE to SE with light gain, filter, then go to high gain. You need to analyze and also look at CM chokes and DE filtering prior to gain possibly. Greater then 1000 gain with a single stage is a little crazy IMO
« Last Edit: May 14, 2020, 03:39:34 pm by coppercone2 »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #4 on: May 14, 2020, 03:44:56 pm »
Looking at http://www.farnell.com/datasheets/2299421.pdf I don't see a gain figure just a maximum of 500µA output.
 

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2431
  • Country: mx
Re: variable gain current amp
« Reply #5 on: May 14, 2020, 04:03:21 pm »
From the datasheet, the output voltage is:
Vout = Rs*Is *Rgain/Rin.
This can be re-written to show the transresistance gain:

A= Vout/Is = Rs*Rgain/Rin

Since Rin is fixed internally, (Rin = 1/Gm, where Gm = 200 uA/V)  you have two options to increase the sensitivity: Increase either Rs or Rgain. Or a combination of both.

Are you planning to switch the ranges with a mechanical or electronic switch?

 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 11341
  • Country: us
  • $
Re: variable gain current amp
« Reply #6 on: May 14, 2020, 04:06:06 pm »
Looking at http://www.farnell.com/datasheets/2299421.pdf I don't see a gain figure just a maximum of 500µA output.

In high voltage operation, it is recommended that IBIAS remainwithin 200 μA to 1 mA. This ensures   that the bias circuit is turned on,  allowing the device to function as expected. At the same time,  the current through the bias circuit/regulator is limited to 1 mA. Refer to Figure 19 and Figure 21 for IBIAS and V+ information when using the AD8212 in a high voltage configuration

Don't know if you can get a guard trace around it.


This is what the LT3080 datasheet says (page 10) about the set pin, which is also high Z
Quote
"With the low level current used to generate the reference voltage, leakage paths to or from the SET pin can create errors in the reference and output voltages. High quality insulation should be used (e.g., Teflon, Kel-F); cleaning of all insulating surfaces to remove fluxes and other resi-dues will probably be required. Surface coating may be necessary to provide a moisture barrier in high humidity environments

Board  leakage  can  be  minimized  by  encircling  the  SET  pin and circuitry with a guard ring operated at a potential close to itself; the guard ring should be tied to the OUT pin. Guarding both sides of the circuit board is required. Bulk leakage reduction depends on the guard ring width. Ten nanoamperes of leakage into or out of the SET pin and associated circuitry creates a 0.1% error in the reference voltage. Leakages of this magnitude, coupled with other sources  of  leakage,  can  cause  significant  offset  voltage  and reference drift, especially over the possible operating temperature range.If  guardring  techniques  are  used,  this  bootstraps  any  stray capacitance at the SET pin. Since the SET pin is a high impedance node, unwanted signals may couple into  the  SET  pin  and  cause  erratic  behavior.  This  will  be  most  noticeable  when  operating  with  minimum  output capacitors at full load current. The easiest way to  remedy  this  is  to  bypass  the  SET  pin  with  a  small  amount  of  capacitance  from  SET  to  ground,  10pF  to  20pF is sufficient."
Thankfully your circuit is well above that (200uA min vs 10uA (3080)typical, but its something to consider because I know you do automotive), which is hot and dirty. I think if you keep dropping bias current, you will get non-linear behavior, and eventually some kind of garbage. That could mean nonlinear noise modulation of gain if its set low. That might be ugly but I am not sure, and maybe with their techniques you can extend the chip functionality to lower bias current. Not a current source but still may be of use.
« Last Edit: May 14, 2020, 04:11:10 pm by coppercone2 »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #7 on: May 14, 2020, 04:35:27 pm »
Well if the maximum output is 500µA surely I just keep cranking up the output resistor until I get the voltage I want that is less than the maximum the chip can output. It's not like I am forcing the current down.

This is a peice of bench equipment. I need to measure quiescent current draw at 50mA and then running current up to 5A, so the probmem is having 50x gain and 5000x gain. I was going to just switch in the different resistors for different gains with a mosfet.
 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 11341
  • Country: us
  • $
Re: variable gain current amp
« Reply #8 on: May 14, 2020, 04:41:35 pm »
also, look at figure 21 in that data sheet, its interesting.

I would try to use a current clamp I think, wide range single range current sensing it hard

You might have to degauss it though, if you go that method
« Last Edit: May 14, 2020, 04:43:45 pm by coppercone2 »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #9 on: May 14, 2020, 05:03:04 pm »
My initial plan was to switch current sense resistors but with a 20mR resistor there is not much that can act as a switch without it being part of the measurement unless i use 300µR power mosfets.
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #10 on: May 14, 2020, 05:22:05 pm »
at a 1mV input the AD error is high. this can't be done with gain adjustments. It has to be in the the sense resistor switching.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: variable gain current amp
« Reply #11 on: May 14, 2020, 07:14:29 pm »
The proposed current sense amplifier is just not very sensitive - more like 2-3 digit class. So it is just the wrong chip for a large dynamic range.  It may run with higher gain, but it does not make too much sense with an input drift of some 7 µV/K and high noise.

For a large range one may have to switch ranges, by switching the shunt, at least in 2 ranges.
20 mOhms are OK for a 5 A range ( 100 mV full scale), but this means 1 mA gives only 20 µV.  So it would need a pretty good amplifier to do this in one range.
For shunt switching on can have the 2 shunts in series. For switching one could use MOSFETs in 2 ways:
a) short out the larger shunt and switch the sense point
b) direct the current with 2 MOSFETs and always sense the voltage with both shunts in series.

If the burden voltage is critical it may need a beefy FET with on ON resistance in the <20 mOhms range.
 

Online Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1478
  • Country: ua
Re: variable gain current amp
« Reply #12 on: May 14, 2020, 07:33:38 pm »
I would use some precision opamp like op37, for example, with a constant gain of 50 (and a 20 mOhm shunt). It has a trim circuit to adjust a voltage offset. The only drawback is that it needs a dual power but it is usual for precision and ultra-precision opamps.
« Last Edit: May 14, 2020, 11:44:17 pm by Vovk_Z »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #13 on: May 14, 2020, 07:48:16 pm »
The proposed current sense amplifier is just not very sensitive - more like 2-3 digit class. So it is just the wrong chip for a large dynamic range.  It may run with higher gain, but it does not make too much sense with an input drift of some 7 µV/K and high noise.

For a large range one may have to switch ranges, by switching the shunt, at least in 2 ranges.
20 mOhms are OK for a 5 A range ( 100 mV full scale), but this means 1 mA gives only 20 µV.  So it would need a pretty good amplifier to do this in one range.
For shunt switching on can have the 2 shunts in series. For switching one could use MOSFETs in 2 ways:
a) short out the larger shunt and switch the sense point
b) direct the current with 2 MOSFETs and always sense the voltage with both shunts in series.

If the burden voltage is critical it may need a beefy FET with on ON resistance in the <20 mOhms range.

Using fets to switch the shunt resistors is not so simple. I need serious fets with µR of resistance to not become a significant and unpredictable part of the measurement. with a 20mR fet I want something like 2 orders of magnitude or better (200µR) lowest i can find are 340µR which really means 680µR when you actually use them, two in parallel will give me a fighting chance or getting something like 200µR.

Given that the solution is in switching the shunts the amplifier becomes pointless as that was all that was for to switch the gains. I can use a 200mV panel meter and  set the decimal point on change over using it at up to 50mV
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: variable gain current amp
« Reply #14 on: May 15, 2020, 12:10:33 pm »
With a 2 nd FET or an analog switch one can avoid having the switch as part of the shunt. So the higher current FET would only need to be so large to keep the overall burden low enough (e.g. 10 mOhms range). The 2 nd FET would only be for the maximum current in the lower range, so less critical. The switching part may still need an auxiliary voltage.

The alternative would be using a really good amplifier for current sense and possibly a slightly larger shunt despite of more heat loss. This would be a little like designing for a high resolution current measurement in the 6 digit range.
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #15 on: May 15, 2020, 12:47:28 pm »
No, you are totally missing the key challenges. I have come up with a solution. One does not simply got a MOSFET that is significantly lower resistance than a current shunt. So I am prepared to loose 100mV. at 5A that is 20mOhms. A 10mOhm MOSFET will introduce a massive error. I have found MOSFETS for over 500A that are 300µOhm. I will use 2 in parallel.

But I need 100mV to look like 5.00 on my voltmeter and I will get 10.00 with 20mohm. If I use a 10mOhm shunt I start to loose accuracy. So I am using my low range shunt as a voltage divider with 2 1Ohm resistors in series and my measurement comes off the centre tap. The combined series and parallel resistances when the higher current shunt is switched in is near bang on 20mOhm still. While the 2Ohms will also drop 100mV at 50mA which being tapped off the middle will be 50mV @ 50mA which I can represent on a 200mV panel meter. The decimal points can be turned on at will so I can make it all work without even needing a current amp.
 

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2766
  • Country: ca
Re: variable gain current amp
« Reply #16 on: May 15, 2020, 01:24:31 pm »
Simon,

Take a look at this:

https://www.analog.com/en/products/ltc6102.html#product-overview

The LTC6102 is zero drift, so it is optimised for wide dynamic range.



Results







Linearity at low currents





The non-linearity is in the input stage, so post amplification will not fix it.

Regards,

Jay_Diddy_B

* 6102 range switch.asc (1.71 kB - downloaded 26 times.)

 
The following users thanked this post: exe, Vovk_Z

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2766
  • Country: ca
Re: variable gain current amp
« Reply #17 on: May 15, 2020, 01:53:11 pm »
Hi,

HP did this in their the 66332A Power supply:



They use two shunts, each one is four terminals. A MOSFET switch shorts the sensitive range shunt. The MOSFET is not in the measurement circuit for the high current range.

You can implement like this:



In this configuration the RDSon doesn't impact the reading. The low range shunt is optimised for low currents.

Regards,
Jay_Diddy_B
 
The following users thanked this post: Vovk_Z

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #18 on: May 15, 2020, 02:05:10 pm »
I am looking at that chip. Yes it has a pretty high gain. Yes if you use two circuits and just bypass the low current one sure it works, but then I'd need to duplicate everything and analogue multiplex to the output to read it.
 

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2766
  • Country: ca
Re: variable gain current amp
« Reply #19 on: May 15, 2020, 02:12:04 pm »
Hi,

I think that double measuring and shorting shunts is the way to get really high dynamic range.

Dave did a review of device designed to this in EEVBlog 1304. Look at 31:00 mins plus. There are multiple shunts and a pair of shorting MOSFETs.

Quitech Otti USB power measurement and battery analyser: https://www.qoitech.com/

The question is how good do you need to be?

Regards,
Jay_Diddy_B


 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #20 on: May 15, 2020, 03:02:00 pm »
I don't need too much accuracy.
 

Online Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1478
  • Country: ua
Re: variable gain current amp
« Reply #21 on: May 15, 2020, 04:54:56 pm »
I have found MOSFETS for over 500A that are 300µOhm. I will use 2 in parallel.
- it seems like a bit no-good design - use two 500 A mosfets to measure 0.05-5 A. What resolution do you need? Look into books like "Opamp application" etc, you definitely find strait way to measure what you need (you have to calculate an error budget).
 

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2766
  • Country: ca
Re: variable gain current amp
« Reply #22 on: May 15, 2020, 05:13:16 pm »
Hi,
How are you measuring the output of the current amplifier?

How many bits do you have?

Can it work close to ground?

Jay_Diddy_B
 

Online Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1478
  • Country: ua
Re: variable gain current amp
« Reply #23 on: May 15, 2020, 05:20:38 pm »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: variable gain current amp
« Reply #24 on: May 15, 2020, 05:33:19 pm »
Hi,
How are you measuring the output of the current amplifier?

How many bits do you have?

Can it work close to ground?

Jay_Diddy_B


it will go to a panel meter. I can get meters that measure 200mV, 2V, 20V. I don't need it to be highly accurate. It's more about repeatability.
« Last Edit: May 18, 2020, 08:45:51 am by Simon »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf