> Bill,
>
> The circuit I\'ve shown is not intended to compare 3 resistors. It is not
> intended
> to measure resistance and it is not intended to replace a Blumlein
> bridge. The sole purpose of this circuit is to measure current
> continuously from 1mA-10A. For measuring current accurately, I need to
> know the exact value of resistors and that needs another setup and as
> you indicated probably a Blumlein bridge.
> -------------------------------------------------------------------------=
Didn't find your answer? Ask the community — no account required.
J
John Larkin
like this.
uses.
ebay and sent to a local
The OPs problem is to measure current, not resistance.
And you still think he should construct his own metrology gear?
It was at the Foothill Electronic Flea Market not because it was junk, but because it was obsolete, displaced by cheap and accurate digital instruments. I got a few for around $10 each. I use them to check our DSP-based LVDT digitizers, and to test electronic wattmeters.
Start up your own primary standards lab and show everyone how to do it right.
Get rid of the home-made diffamp and use a CMOS 24-bit delta-sigma adc, AD7793 maybe. They have differential inputs that are essentially perfect, programmable gain ranges, nanovolt drift, and are linear to a few PPM. And add a signal-level mux to get the switch resistances out of the measurement path. And get some good resistors. And you're done.
The frequency one should be extra interesting. Accurate machining, glass blowing, filling the cell .... ooops you have to measure the buffer pressure .... making an accurate pressure measuring device, obtaining the Laser, biasing the Laser .... ooops got to measure a current, making the meter to measure the current, building the VCXO, .... ooops got to measure the frequency to test the VCXO, making a frequency counter .... ooops the frequency counter needs a reference frequency, making a primary frequency standard, Accurate machining, glass blowing, filling the cell .... ooops you have to measure the buffer pressure .... making an accurate pressure measuring device, obtaining ............ and so on for ever.
M
MooseFET
On Jun 26, 7:12=A0am, John Larkin wrote: [... stuff then diffamp ...]
Better yet, run the whole analog and ADC process on an isolated supply that has a point on the measurement circuit as its common. This way, the common mode is only equal to a little drop in some wiring.
J
John Larkin
Yup, it's not hard to float a serial adc. Analog Devices will sell you a single part that will provided isolated power and isolate the data.
This floats 16 channels of d-s ADC. We used cheap IDSN transformers as the power isolators, and used the inverter frequency to clock the ADCs.
formatting link
After a year or so, channels are still accurate to single digits of PPM.
A lot of stuff that once was a whole bench full of metrology gear, that took days to make a measurement with, can now be encapsulated into a $6 delta-sigma ADC with built-in reference.
I once saw the premium General Radio frequency measurement system ca
1950; it was mostly unused and dusty when I did see it. It was one whole wall of a room, six rack bays maybe, full of ovenized crystal and tuning fork oscillators, interpolation oscillators, lissajous scopes, mixers, multipliers, tunable filters, all sorts of hard-to-use stuff. All to do the work of a $99 frequency counter.
John
R
Rich Grise
It's worse than that - if you buy a bushel of fruit from a vendor, what guarantee is there that you're both using the same bushel?
Well, we've all known for some time that Sloman's an idiot.
Why are you switching the resistors at both the top and bottom? (Perhaps there is something I'm missing.) Seems like you could remove the switches on the ground side. Then see if this reduces your error. Your problem might be just switch resistance as others have suggested.
George H.
J
JosephKK
-
his
suggestions like this.
other uses.
No. They do not. I had coworkers that worked in a primary standards lab some 25 years ago. Far worse, many can easily be damaged beyond repair, and the real stuff is expen$$$$ive.
on ebay and sent to a local
Perhaps OP is insufficiently skilled in the methods and uses?
Which is usually intended to be operated in a temperature controlled bath. SOP for standard resistors. Used them versus a test resistor in with a 7 digit Kelvin-Varley divider with a decent voltage reference and a good DMM measuring the difference. Had about 100 parts to test and had to find means to switch them into the measurement circuit and back to the maintenance circuit, one by one, sequentially.
I doubt that your contact resistances are that low, nor are they repeatable enough to get the results you are looking for. Redo the circuit so that you are not measuring the voltage across the current path relay/switch contacts. The inconsistent contact resistance will introduce nearly 1 % errors with the lowest current shunt value (highest range).
------------------------------------------------------------------------------------------ Ok,I will do it, If I want to remove switches then I have to place one Diff. Opamp for each range and use an analog multiplexer. and then I have different Offset/Gain/CMRR errors for each range that have to be calibrated. While I doubt that this problem goes away with just removing switches but it worth a try. this problem of reading different values for each range exists even when current is just 100mA. I think when current is high switches may create significant non-linearity but because I have this problem even in lower currents and even when I used
0.1% metal film resistors I think I more need a good calibration procedure. I also tended to try linnix's suggestion of creating look up tables but interestingly no one else here endorsed his suggestion.
24-bit delta sigma A/D converters don't really do better than 20-bits in practice. They rely on pulse width modulation to get roughly one part per million precision, and if you dig into the details it gets difficult to do better.
That why I like ratio transformers, which don't start falling over until you get to one part in ten million (one part per billion if you are prepared to do it all in binary).
The national laboratory bridge systems don't really take advantage of op-amps the way they should. I'd like to put a system together where excitation current never flows through the precision windings - op amps would monitor the voltages across the precision windings and drive the appropriate excitation current through slapped on secondary windings (with suitable precautions to prevent op-amp offsets building up direct current in the driven windings).
I once swapped out a bunch of dry-reed relay switches - gold-on-gold in a sealed tube - and replaced them with mercury wetted reed relays, because the 75 milliohm contact restance of the mercury-wetted reed relays was not only lower that of the dry-reed relays, but also tolerably stable.
I got to be able to do it because I was able to show that the erratic contact resistance of the dry-reed relay had wrecked the settling time of the circuit that I was workingv on - we trimmed the settling on each range to be dead-beat with a separate trimming potentiometer, and they didn't stay dead-beat until I swapped in the mercury wetted reeds.
-- Bill Sloman, Nijmegen
B
Bill Sloman
o -
his
ng
ons like this.
her uses.
on ebay and sent to a local
.
M
r
DC
Which weren't as good, but good enough for the application, and easier to use.
Lucky you. If you'd really appreciated the bargain that you'd got. you might have found other applications
That would cost more money than I want to spend - in the absence of obvious customers - and I share your pessimism when it comes to drumming up business for an independent calibration lab.
In fact I could probably assemble a cheap Blumlein bridge for very little money, but that is a very small proportion of a calibration laboratory. A Thompson-Lampard calculable capacitor is probably cheap as primary standards go, but it isn't a primary standard without a $20,000 laser interferometer, which is one of several reason why I discouraged Petkovic's enthusiasm for the device.
If you want to make satrical suggestions, try to go for propositions that aren't totally absurd.
-- Bill Sloman, Nijmegen
M
MooseFET
The OP will be working with fairly low voltages. He is also workin at low frequencies. He may gain some advantage from a flying capacitor voltage multiplier before the ADC.
If charge injection doesn't trash the accuracy. That becomes another tricky design issue which I'm guessing he doesn't need.
A 24-bit delta-sigma is a huge amount of precision, already worked out by experts, for sale cheap.
John
B
Baron
Lookup tables are commonly used in industrial instrumentation !
Best Regards:
Baron.
B
Bill Sloman
Nobody is saying that you can't damage the equipment that measures the physical effects that now serve as primary standards for everything except the kilogram. What you can't do is get it to give you a wrong answer when it is working right.
The kilogram is eventually going to go the same way
formatting link
Some 25 years ago the process wasn't so far advanced.
The resistors that might be candidates for use as secondary standards have temperature coefficients around the 1ppm/C level.
You'd probably use them in a temperature-controlled environment, but while a temperature contolled bath could be set up to stabilise the local temerature to a micro-degree or so, the process of measuring resistance involves dissipating heat within the resistor, which makes heroic temperature stabilisation a little pointless.
And a 7-digit Kelvin-Varley divider used with a decent direct current voltage source is .obsolete technolgy in this context.
Modern sceme for comparing resistors use coaxial alternating current bridges built with ratio transformers.
-- Bill Sloman, Nijmegen
B
Bill Sloman
-
s
Standards authorities. This has been obviously necessary since markets were invented. We've got evidence that people were setting local standards some 4000 years ago (in the valley of the Indus) .
formatting link
Standards
and the process has been going on ever since.
The point is that we now have a number of independent national standards authorities who all trade secondary standards to maintain a continuous check on the quality of their work. A non-national standards lab would have to be just another partner in this process, which would make their results just as traceable as anbody else's.
Rich does seem to know a lot of things that don't happen to be true - he's not only ignorant, but also unaware of the true depth and extent of his ignorance.
-- Bill Sloman, Nijmegen
Join the Discussion
Have something to add? Share your thoughts — no account required.
Didn't find your answer?
Ask the community — no account required
Report Content
You are reporting this content to the moderators. They will look at it
ASAP.