Hi all,
I'd like to be able so see voltages down to the microvolt on my scope (Tektronix 2232, max sensitivity is 2mV/div), to measure low noise power supplies, DACs etc...
I know some old Tek scopes can measure down to the microvolt level, but that could cost me more than 200€
Is a low noise preamplifier possible? In DIY too?
I'd like to have some advices.
Thank you, and merry christmas to all
Alex
I'd like to be able so see voltages down to the microvolt on my scope (Tektronix 2232, max sensitivity is 2mV/div), to measure low noise power supplies, DACs etc...
I know some old Tek scopes can measure down to the microvolt level, but that could cost me more than 200€
Is a low noise preamplifier possible? In DIY too?
I'd like to have some advices.
Thank you, and merry christmas to all
Alex
OK
I'll try to contact John Curl, and Scott Wurcer to see what they think.
I was also thinking about the 797, but strangely, in the dhatasheet, they recommand other opamps for sources with high impedance (AFAIK, a probe is >100k)
I think the layout and PS won't be trivial for this
Thanks Elso!
I'll try to contact John Curl, and Scott Wurcer to see what they think.
I was also thinking about the 797, but strangely, in the dhatasheet, they recommand other opamps for sources with high impedance (AFAIK, a probe is >100k)
I think the layout and PS won't be trivial for this
Thanks Elso!
Hi Bricolo,
I found a schematic of the AC millivoltmeter in an old Elektor issue that you may use as an instrumentation amp. Its most sensitive range is 200uV, which could be just about enough for your purposes. The design itself is quite simple (FET at the input, discrete opamp). The article and the schematic can also be found in "Electronic Test Instruments" by Elektor Electronics, 1994 (ISBN 0 905705 37 8).
Regards,
Milan
I found a schematic of the AC millivoltmeter in an old Elektor issue that you may use as an instrumentation amp. Its most sensitive range is 200uV, which could be just about enough for your purposes. The design itself is quite simple (FET at the input, discrete opamp). The article and the schematic can also be found in "Electronic Test Instruments" by Elektor Electronics, 1994 (ISBN 0 905705 37 8).
Regards,
Milan
Bricolo, first you must specify how low the noise signal is that you want to show.
Let's say you want to see 10 uV, then the preamp must have 1 uV input noise or less.
If you take an opamp with less nosie than 4 nV/Hz you will have a input noise 0.6 uV or so.
A simple opamp like NE5534 may very well fit your needs but if you need even lowever noise maybe an LT1115, LT1028 or AD797 will do the job.
Connect the opamp in non-inverting mode and make sure the the feedback has pretty low values, less than 100 ohms from inverting input down to ground. Feed the opamp with batteries, 2 x 9 volts or more with good decoupling. Don't förget to place the whole thing is metal box.
Since it's a measuring gear I would have both input and output coupling caps. You must probably also have DC-gain 1 or use a DC-servo. Only a few opamps will manage DC gain of 1000.
I really recommend that you check the datasheets of LT1028/LT1115/AD797
Other useful opamps could be OPA627 or more preferable OPA637. They haven't got extremely noise noise but pretty low.
OP27 or more preferable OPA37.
Don't forget to limit the very low frequency nosie below 20 Hz.
I must warn you that it's pretty hard to see uV without interference of hum and other interferences.
Let's say you want to see 10 uV, then the preamp must have 1 uV input noise or less.
If you take an opamp with less nosie than 4 nV/Hz you will have a input noise 0.6 uV or so.
A simple opamp like NE5534 may very well fit your needs but if you need even lowever noise maybe an LT1115, LT1028 or AD797 will do the job.
Connect the opamp in non-inverting mode and make sure the the feedback has pretty low values, less than 100 ohms from inverting input down to ground. Feed the opamp with batteries, 2 x 9 volts or more with good decoupling. Don't förget to place the whole thing is metal box.
Since it's a measuring gear I would have both input and output coupling caps. You must probably also have DC-gain 1 or use a DC-servo. Only a few opamps will manage DC gain of 1000.
I really recommend that you check the datasheets of LT1028/LT1115/AD797
Other useful opamps could be OPA627 or more preferable OPA637. They haven't got extremely noise noise but pretty low.
OP27 or more preferable OPA37.
Don't forget to limit the very low frequency nosie below 20 Hz.
I must warn you that it's pretty hard to see uV without interference of hum and other interferences.
I haven't sent the boards out yet, but the SSM2019 works beautifully as a differential amplifier for testing, I measured the THD at 0.0008% to 0.0014% -- running it from a Lambda laboratory supply. I had intended to use the SSM2017 which Walt used in the second of the series of articles on low-noise regulators in Audio Electronics circa 1995, but couldn't find the chip. The circuit can be made simpler by omitting the supply and just using a pair of 9V batteries:
gain is selectable, 20, 40 and 60dB -- there are two series of resistors so that the gain can be precisely set.
An externally hosted image should be here but it was not working when we last tested it.
gain is selectable, 20, 40 and 60dB -- there are two series of resistors so that the gain can be precisely set.
Bricolo said:
certainly, but differential is very nice to have -- the circuit above has subsequently been modified to provide some clamp diodes on the input to protect the opamp --
btw, the SSM2019 does run a little "hot" -- and mine consumed a lot more gas than Analog Devices described on their term-sheet, there was no oscillation eating watts...i would suggest that a cementable heat sink would help the chip.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.