Electronics > Projects, Designs, and Technical Stuff

New favourite opamp for measuring equipment - AD8605ARTZ

(1/4) > >>

ddavidebor:
Hello guys,

I want to share with you a component that I have decided to adopt in a broad range of products (scientific equipment for physics).

I was looking for an opamp with as many of the following characteristics as possible:


* RRIO (Rail to Rail Input and Output)
* offset voltage under 100uV max 50uV typical
* input bias current < 100pA
* output current >40mA with 1V drop
* unity gain stable
* stable with capacitive loads
* continuous output short circuit without damage
* cheap to source in Shenzhen
* available in single packages
* under 40 cents @ 100pcs
* second source available
The target product is a wide variety of measuring equipment where thus opamp should be used in a series of standard analog circuitry, such as:


* input buffers
* precision comparators
* inverting and non-inverting amplifiers
* filters
* precision rectifiers
* voltage clamps
* ac amplifiers
* current sources
Now, I did not think i would find something like this... but indeed there is, the AD8605 and it's Chinese pin compatible equivalent TP2311 available for around 0.3$ and 0.22$ respectively

The technology Analog Devices used to achieve such performance is pretty interesting https://www.analog.com/en/design-center/landing-pages/001/digitrim-technology.html

This little buddy does the work of laser-trimmed opamps at a much lower price

Kleinstein:
The data of this OP don't look very impressive.  In the graph the CMRR looks good at high frequency - not sure if this is true.
The offset voltage as a function of common mode voltage looks relatively poor, indicating poor CMRR (some 70dB) even at low frequency.

Rail to Rail input is a two sided thing: they usually have a cross over region, where the input changes between input stages. So in this case there is a range of some 0.5 V (e.g. 3.3-3.8 V with 5 V supply) that is no usable for precision applications.

OM222O:
The high output current (40mA is on the large side for an op amp!) contradicts the precision requirements.
Also no bandwidth limit was mentioned  ??? A lot of physics instruments run at high frequencies  :-//
The capacitive load requirement is also odd ... non of the applications you listed require high capacitive loads.

anyhow, my personal go to op amps for precision work are the MAX4238 and the OPA188 / OPA187. they are a bit pricy (>1$ even at volume) but have been very decent and robust parts.

ddavidebor:

--- Quote from: Kleinstein on September 04, 2019, 12:38:02 pm ---The data of this OP don't look very impressive.  In the graph the CMRR looks good at high frequency - not sure if this is true.
The offset voltage as a function of common mode voltage looks relatively poor, indicating poor CMRR (some 70dB) even at low frequency.

Rail to Rail input is a two sided thing: they usually have a cross over region, where the input changes between input stages. So in this case there is a range of some 0.5 V (e.g. 3.3-3.8 V with 5 V supply) that is no usable for precision applications.

--- End quote ---

The cross-over region is an excellent point. I was not able to get it to act out in simulations, do you have a suggestion on how i would be able to better test this behaviour?

ddavidebor:

--- Quote from: OM222O on September 04, 2019, 01:03:21 pm ---The high output current (40mA is on the large side for an op amp!) contradicts the precision requirements.
Also no bandwidth limit was mentioned  ??? A lot of physics instruments run at high frequencies  :-//
The capacitive load requirement is also odd ... non of the applications you listed require high capacitive loads.

anyhow, my personal go to op amps for precision work are the MAX4238 and the OPA188 / OPA187. they are a bit pricy (>1$ even at volume) but have been very decent and robust parts.

--- End quote ---

In my case bandwidth requirements are very limited. Many of the instruments I design stop at a few hundred KHz. Not really in the counting photons business for now (though I do have one light speed measuring apparatus in the work that works at 2GHz :-) - well, will I hope. )

For the capacitive loads... peak detectors have a "hold" capacitor inside. When using it as an analog output buffer (I often have many analog outputs as everyone prefers a different datalogger and big resellers like to drop their favourite one in the kit as well) and there is the capacitive load of the wire and possibly datalogger input.

I also like and have used the MAX4238 as my go-to chopper opamp.

High output current has been very useful to me to simplify the feedback loop of current generators by removing the mosfet/bjt. Yeah i guess i could use a different opamp for that, and indeed had. Trying to slim the bom helps the wallet with the more low production number equipment.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod