Spice model CA3080 and/or LM3080 
Author Message
 Spice model CA3080 and/or LM3080

Hi,

I'm looking for a Spice model of the CA3080 and/or the LM3080. Can somebody
help me?!??!

Thanks!

Paul

--
----------------------------------------------------------------------------
-------------------------
LET OP!!!
Per 3 aug. as. is ons adres:
----------------------------------------------------------------------------
-------------------------
Salland Electronics bv
Katwolderweg 15
8041 AC  Zwolle

Tel.: +31 (0)38-454 0864
Fax.: +31 (0)38-423 8160
Web: http://www.***.com/

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



Tue, 16 Mar 2004 16:27:40 GMT
 Spice model CA3080 and/or LM3080
I don't have a SPICE model, but I do want to use a '3080 in a new design (can't
use the later LM13600 dual because of low supply volts). Can anyone suggest good
sources of '3080?


Tue, 16 Mar 2004 18:00:26 GMT
 Spice model CA3080 and/or LM3080

Quote:

> I don't have a SPICE model, but I do want to use a '3080
> in a new design (can't use the later LM13600 dual because
> of low supply volts). Can anyone suggest good sources of
> '3080?

 A. Any Intersil distributor.  This was originally an RCA
 design, which was then 2nd-sourced by NSC.

 RCA was purchased by Harris.  Intersil was also purchased
 by Harris as was the GE semiconductor operation.  After
 many years passed, whatever was left of this conglomeration
 was spun off by Harris a few years back to an investor
 group, who called themselves - DRUMBEAT - Intersil.

 Intersil has been very good about keeping old designs in
 production, including the venerable CA3080.  Three cheers!
 What other part can operate reasonably at 2.5V and 50nA?

 A search at this Intersil URL shows all the CA3080 models
 in production, http://www.intersil.com/design/index.asp

 They have a new April 2001 datasheet, showing off (fig. 30)
 the device's low 0.023pF input-to-output capacitance!  :-)

 http://www.intersil.com/data/FN/FN4/FN475/FN475.pdf

 The CA3080 is a great chip, but it has one serious problem,
 the Iabc current mirror.  This is not a Wilson mirror, as it
 should be (see AoE page 89), and it's therefore subject to
 the Early effect, which results in a changing mirrored Iabc'
 with any change in the input-pair common-mode voltage, which
 damages the chip's accuracy.  This is important, because Iabc'
 programs the 3080's basic property, its transconductance,
 g_m = qI/2kT   Sadly they don't document this problem in the
 datasheet performance curves.  Nor did NSC in their datasheet,
 http://www.freetradezone.com/design_center/images/pdf/D0083250.pdf

 Intersil has a discussion of the issue, along with dramatic
 curve-tracer photos, on pages 16 and 17 of app note an6668,
 http://www.intersil.com/data/an/an6/an6668/an6668.pdf
 The Early voltage can be estimated from fig 39D at about 25V,
 and if the transistors are the same as in the '3080, this can
 be used in creating a reasonable Spice model (the rest of the
 info needed is in the datasheet).

 NSC fixed this problem in their LM13700, which is still
 available, see http://www.national.com/ds/LM/LM13700.pdf
 This part can operate reasonably at 3V and 50nA.   :-)

 Thanks,
          - Win

 Winfield Hill
 Rowland Institute for Science
 100 Edwin Land Blvd
 Cambridge, MA 02142-1297



Tue, 16 Mar 2004 20:35:02 GMT
 Spice model CA3080 and/or LM3080
My sincere thanks - The answer I asked for, PLUS the answers to the questions I
didn't know about yet. My current '3080 use is an inverting amplifier, so the
Early effect shouldn't be a problem to me. I've downloaded the new data sheet.

I'll also look again at the LM13700 on your suggestion - I had considered the
LM13600, but it had seemed to me at the time that the extra complexity in the
better current mirrors would rule it out at low voltage, at least for my
application.

The biggest surprise for me is the ability to work under 100nA, even at 125C.
I've tried unsuccessfully in the past to get down into this region; I think I'll
resurrect that project too.

Regards,
Tony Roe

Quote:


>> I don't have a SPICE model, but I do want to use a '3080
>> in a new design (can't use the later LM13600 dual because
>> of low supply volts). Can anyone suggest good sources of
>> '3080?

> A. Any Intersil distributor.  This was originally an RCA
> design, which was then 2nd-sourced by NSC.

> RCA was purchased by Harris.  Intersil was also purchased
> by Harris as was the GE semiconductor operation.  After
> many years passed, whatever was left of this conglomeration
> was spun off by Harris a few years back to an investor
> group, who called themselves - DRUMBEAT - Intersil.

> Intersil has been very good about keeping old designs in
> production, including the venerable CA3080.  Three cheers!
> What other part can operate reasonably at 2.5V and 50nA?

> A search at this Intersil URL shows all the CA3080 models
> in production, http://www.intersil.com/design/index.asp

> They have a new April 2001 datasheet, showing off (fig. 30)
> the device's low 0.023pF input-to-output capacitance!  :-)

> http://www.intersil.com/data/FN/FN4/FN475/FN475.pdf

> The CA3080 is a great chip, but it has one serious problem,
> the Iabc current mirror.  This is not a Wilson mirror, as it
> should be (see AoE page 89), and it's therefore subject to
> the Early effect, which results in a changing mirrored Iabc'
> with any change in the input-pair common-mode voltage, which
> damages the chip's accuracy.  This is important, because Iabc'
> programs the 3080's basic property, its transconductance,
> g_m = qI/2kT   Sadly they don't document this problem in the
> datasheet performance curves.  Nor did NSC in their datasheet,
> http://www.freetradezone.com/design_center/images/pdf/D0083250.pdf

> Intersil has a discussion of the issue, along with dramatic
> curve-tracer photos, on pages 16 and 17 of app note an6668,
> http://www.intersil.com/data/an/an6/an6668/an6668.pdf
> The Early voltage can be estimated from fig 39D at about 25V,
> and if the transistors are the same as in the '3080, this can
> be used in creating a reasonable Spice model (the rest of the
> info needed is in the datasheet).

> NSC fixed this problem in their LM13700, which is still
> available, see http://www.national.com/ds/LM/LM13700.pdf
> This part can operate reasonably at 3V and 50nA.   :-)

> Thanks,
>          - Win

> Winfield Hill
> Rowland Institute for Science
> 100 Edwin Land Blvd
> Cambridge, MA 02142-1297



Wed, 17 Mar 2004 07:25:14 GMT
 
 [ 4 post ] 

 Relevant Pages 

1. Spice model CA3080 and/or LM3080

2. Need SPICE models: LM3080 (OTA) & MPF102 (JFET)

3. Need SPICE models: LM3080 (OTA) & MPF102 (JFET)

4. Ich brauche die protel spice 3 models für CA3080 und CA3140

5. I need the protel spice models für CA3080 and CA3140

6. Ich brauche die protel spice 3 models für CA3080 und CA3140

7. I need the protel spice models für CA3080 and CA3140

8. NTE2402 NPN Spice Model (or NTE Spice Library)

9. CA3080 (OTA) replacement?

10. Cloning the ca3080

11. Wanted: SPICE device model parameters

12. Spice models for thyristors


 
Powered by phpBB® Forum Software