Voltage compliance of the programmable current source with REF200

[Jiaqi]

I am designing a low current source (decades to hundreds of
microamps) and I found Burr-Brown's REF200 satisfying. I read BB's
application bulletin AB-061 "IMPLEMENTATION AND APPLICATIONS OF CURRENT
SOURCES AND CURRENT RECEIVERS", in which Fig. 14 shows a programmable
current source consisting of two resistors, an REF200 and an external
op amp. It says in the text,
"Because compliance of the 100uA current source is 2.5V, the
current source, Figure 14, can only COMPLY WITHIN 2.5V OF THE NEGATIVE
RAIL-even if the op amp can go further."
This sentence is above my comprehension. I read the datasheet of
REF200 but cannot yet find any clue. As far as I know, the so-called
"voltage compliance" refers to the valid range of the load terminal
potential. It seems to me that the load terminal potential has to be at
least 2.5V HIGHER than the negative rail, which contradicts the
original text.
Any instruction or information is welcome.

 

[[email protected]]

I am designing a low current source (decades to hundreds of
microamps) and I found Burr-Brown's REF200 satisfying. I read BB's
application bulletin AB-061 "IMPLEMENTATION AND APPLICATIONS OF CURRENT
SOURCES AND CURRENT RECEIVERS", in which Fig. 14 shows a programmable
current source consisting of two resistors, an REF200 and an external
op amp. It says in the text,
"Because compliance of the 100uA current source is 2.5V, the
current source, Figure 14, can only COMPLY WITHIN 2.5V OF THE NEGATIVE
RAIL-even if the op amp can go further."

http://focus.ti.com/lit/ds/symlink/ref200.pdf

bottom of page five, left ahnd column.

This sentence is above my comprehension. I read the datasheet of
REF200 but cannot yet find any clue. As far as I know, the so-called
"voltage compliance" refers to the valid range of the load terminal
potential. It seems to me that the load terminal potential has to be at
least 2.5V HIGHER than the negative rail, which contradicts the
original text.
Any instruction or information is welcome.

You are right - the sentence is nonsense. What the author presumably
meant to say is that the current source will only deliver the rated
100uA for loads presenting a voltage at least 2.5V above the negative
rail,

This is determined by the performance of the REF200 - if you look at
page three of the data sheet

http://focus.ti.com/lit/ds/symlink/ref200.pdf

there a two currves for "currrent source - output current versus
voltage". The one on the right shows the performance in the range 0V
to 5V.

Application notes are written by fallible human beings. Even Jim
Williams at Linear Technology is not always 100% right - though he
comes pretty close.

 

[Jiaqi]

This is determined by the performance of the REF200 - if you look at
page three of the data sheet

http://focus.ti.com/lit/ds/symlink/ref200.pdf

there a two currves for "currrent source - output current versus
voltage". The one on the right shows the performance in the range 0V
to 5V.

Thank you very much. I agree that LT's Jim Williams is a great man. I
enjoy several technical documents written by him.
Regards,
Jiaqi Shen

 

[Winfield Hill]

Jiaqi wrote...

Thank you very much. I agree that LT's Jim Williams is a great
man. I enjoy several technical documents written by him.

Me too. As for the REF200, it's an interesting part,
but note TI's detailed information about its status
in the "Pricing/Packaging/CAD Design Tools/Samples"
and "Inventory" sections of the REF200's web page,
http://focus.ti.com/docs/prod/folders/print/ref200.html

Here you learn the miniDIP package version has been
discontinued. This happened long ago, and only the
soic-8 package is still offered. That's unusual and
interesting information for a start. You also learn
there are NO factory REF200 parts in stock and NO parts
in progress, which would also be unusual for an "active
status" part. Finally you learn that anyone ordering
enough parts to trigger a factory production run (I
wonder how many that is) would have to wait 20 weeks.

TI says Arrow Southern Europe, whatever that is, has
291 parts in stock, but I wasn't able to track them down.

The REF200, Jiaqi, is basically dead. Too hard to make,
with it's laser trimming, too impractical with it's
three separate laser-trimmed circuits, and too easy to
replace with cheaper alternatives. R.I.P.

So, as an aspiring engineer, what you learn is that in
"designing a low current source (decades to hundreds
of microamps)" you should NOT find Burr-Brown's REF200
"satisfying," but rather simply note the concepts they
present, and seek other ways to solve your problem.

For example, you might consider the technique shown
in AB165, figures 15A and 15B. This techniques uses
a voltage reference with an adjustable resistor, which
makes more practical sense than using a current-source
reference with an adjustable resistor.

BTW, where did you find the AB-061 app note? I found
your circuit figure 14, and the same description, in
AB165, a 30-page app note.
http://focus.ti.com/analog/docs/techdocsabstract.tsp?abstractName=sboa046

 

[Fred]

I have found recently that several "Burr Brown" parts have hit the 20
week leadtime. Scuttlebutt has it that there were several lines
transfered from the US to overseas. For all intents and purposes it is
dead. In this day and age ( 90% of the time) if the part isn't listed
in stock at Digikey or Mouser most engineers won't design it into a
product.
Fred

 

[Fred]

I have found recently that several "Burr Brown" parts have hit the 20
week leadtime. Scuttlebutt has it that there were several lines
transfered from the US to overseas. For all intents and purposes it is
dead. In this day and age ( 90% of the time) if the part isn't listed
in stock at Digikey or Mouser most engineers won't design it into a
product.
Fred