Capacitive Voltage divider, for O-scopes (10,000:1) / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Capacitive Voltage divider, for O-scopes (10,000:1)

first previous 7 8 9 10 next last

Move Thread

LAN_403

Proud Mary

Wed Jul 14 2010, 11:21AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Fabulous insulator, Patrick.

I wonder what its dielectric constant may be.

I meant to ask you, how vulnerable d'you think your rig might be to unwelcome stray capacitance, and its frequency-related effects ? Can you null it out with a 'set-zero' type feature?

IntraWinding

Wed Jul 14 2010, 01:27PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

I definitely want to have a go at making insulators next time I get a chance. Do they use special clays and glazes used for insulators, or is it the same stuff used for pots?

-

I think I've spotted a circuit problem:

Referring to the data sheet

, at +/-15V supply the TL082CP has an 'Input Common-Mode Voltage Range' of only +/-11V guaranteed. Exceed that and anything could happen!

I'm not sure why your solution of reducing R6 from 10M to 1M improves the simulated response, but it doesn't seem like a good solution to the problem to me, unless I'm missing something?

To me the solution seems to be to increase C2 (perhaps double it?) to get a lower voltage at the op amp input and increase the op amp gain to bring the overall ratio back to 1000.

EDIT: ....And then you can also leave R6 at 10M too, which helps with the low frequency response.

Patrick

Wed Jul 14 2010, 03:01PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

This is the most current schematic yet.

the changes are as follows: R6 has been reduced from 10M to 1M ohms, the pot 100k has been added, with a 47k in series. the max lower feedback resistor is 147k ohms, the min is 47k ohms. thus the gain can be varied across 1.07 to 1.21. (midpoint gain is 1.143 @70 kohms) the R6 was changed due too saturation at the + input causing a 400mV offset above zero( @ 11V signal in ), after it was chaged to 1M the offset dropped to 3-6mV which is probably inherent to the TLO82CP.

now for the divison ratio, first by the op-amp gain limit of 1.07 to1.21:
the max ratio is 12,142 @ c2=8.5nF
the min ratio is 10,714 @ c2=7.5 nF
the most optimal ratio is c2 being equal to 7.8 to 8.0 nF ( 11,142 and 11428 respectively )

so 100,000 V / 11,142 = 8.750 V x 1.1423 G = 10.000v
or 100,000 V / 11,428 = 8.975 V x 1.1142 G = 10.000v

----------------------------------------- -------------------------------------------------- --

Proud Mary wrote ...

I meant to ask you, how vulnerable d'you think your rig might be to unwelcome stray capacitance, and its frequency-related effects ? Can you null it out with a 'set-zero' type feature?

thats what the trimpot is for, also do you think a trim pot is stable for longterm use? or should it be replaced with a fixed value resistor?

IntraWinding wrote ...

I definitely want to have a go at making insulators next time I get a chance. Do they use special clays and glazes used for insulators, or is it the same stuff used for pots?

well i use standard clay for the arts but id like to get real electrical intended stuff.

IntraWinding wrote ...

I think I've spotted a circuit problem:

Referring to the data sheet

, at +/-15V supply the TL082CP has an 'Input Common-Mode Voltage Range' of only +/-11V guaranteed. Exceed that and anything could happen!

ahhh! crap, did i overlook that ?! EDIT: your right, its limited to +11, -11, so +110kv to -110kv. But this will do. for now

IntraWinding wrote ...

I'm not sure why your solution of reducing R6 from 10M to 1M improves the simulated response, but it doesn't seem like a good solution to the problem to me, unless I'm missing something?

To me the solution seems to be to increase C2 (perhaps double it?) to get a lower voltage at the op amp input and increase the op amp gain to bring the overall ratio back to 1000.

let me simulate that and lets see.

Proud Mary

Wed Jul 14 2010, 04:42PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Is the trimmer intended to be a multi-turn type? Do you anticipate needing to crank back and forth across the whole of the 100K, or could you narrow it down with a fixed resistor, and then have a much lower value multi-turn pot to bracket the working resistance range?

Patrick

Wed Jul 14 2010, 05:37PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

i intended to use a singleturn trim pot as 100k was the smallest bracket (1.07-1.2) i thought i could get away with. And it does need to traverse most of that 100k, i think. are those 15-turn little blue rectangle boxes more accurate?

Proud Mary

Wed Jul 14 2010, 07:01PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Patrick wrote ...

i intended to use a singleturn trim pot as 100k was the smallest bracket (1.07-1.2) i thought i could get away with. And it does need to traverse most of that 100k, i think. are those 15-turn little blue rectangle boxes more accurate?

I know only of three, five and ten-turn trim pots, but clearly if you have ten or fifteen entire rotations of 360 deg, then a resistance can be very much more accurately set than with 270 or so degrees in just one single rotation. But if your set-zero is to be a front panel control, you should be looking at ten-turn pots with a vernier drive rather than at a trimmer, and perhaps introduce a temperature compensating element too.

IntraWinding

Wed Jul 14 2010, 09:13PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

I'm not sure if they use special electrical clay or glazes, but presumably some are better than others. Perhaps there's some knowledge out there, anyone?

Patrick

Wed Jul 14 2010, 10:03PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Proud Mary wrote ...

ten-turn trim pots, but clearly if you have ten or fifteen entire rotations of 360 deg, then a resistance can be very much more accurately set than with 270 or so degrees in just one single rotation. But if your set-zero is to be a front panel control, you should be looking at ten-turn pots with a vernier drive rather than at a trimmer, and perhaps introduce a temperature compensating element too.

well for the first of 8 i will just build a rudimentary prototype to pin down the unkowns (one with no external knob). i realize that more turns is more accurate, but what about the wiper contact, will it be stable long and short term? or should i use a collection of fixed value resistors?

10K trim pot, 15 turns cermet type.

Proud Mary

Wed Jul 14 2010, 11:32PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Patrick wrote ...

10K trim pot, 15 turns cermet type.

I'm afraid I don't really know enough to be that much use on this one, Patrick, but I'd assume that a multi-turn trimpot from a manufacturer such as Bourns would be dependable for most purposes.

Patrick

Thu Jul 15 2010, 12:22AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

ok , well i noticed that you used 4 fixed reistors for the lower limb of your divider but you have to be careful that the wiper or one of the lower resistor dont open, thus becoming a 75kv hazard at the low end. i dont have that as a possiblity so i will try that trim pot then. we will see.

first previous 7 8 9 10 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob