On Wed, Mar 12, 2025 at 07:59 PM, DerekK wrote:
BTW, the LM121 was not a chopper stabilized version.
Yup.
?
It effectively competes, having performance that chopper stabilized amps could achieve.
Imagine expecting to see a set gain and come to find that the 200k resistor measures 300k!
Ouch!
?
It keeps us on our toes.? :-)
?
Andy
?
|
On 3/12/25 6:31 PM, DerekK wrote: True Andy. I meant to refer to the LM108 in the app note. The circuit I have is using the LM118 though. Still looking for that LM121 model. That old LM121 is basically just a differential pair. Plus some other details of course. But no voltage gain or output stage as would be found in an opamp. -- David Schultz "The cheeper the crook, the gaudier the patter." - Sam Spade
|
Andy,?
I deal with a lot of old circuitry. Mica caps going bad, as you stated, are VERY rare. It does not seem to be a DC problem as I am feeding in a square wave at 25kHz. I doubt over voltage has ever hit the circuit. PCB has conformal coating and, after swapping the opamp set, still functions the same way.
?
BTW, the LM121 was not a chopper stabilized version. The ap-note clearly states that the LM121 keeps the noise level down because it is nota chopper-type.?
?
I have a scheme to pull the opamps and substitute a different opamp that has low drift and reasonable GBW. I am thinking it may be resistors as well, but initial in-circuit checks show nothing out of the ordinary. I have seen resistors drift to become >+20% out of tolerance due to aging. Imagine expecting to see a set gain and come to find that the 200k resistor measures 300k!
|
On Wed, Mar 12, 2025 at 07:31 PM, DerekK wrote:
The issue is that the circuit I have is not bench performing the way one would expect. Thus I am wondering if it is possibly the mica caps that are use; though I have not heard of mica caps going bad as often as tantalum and electrolytics.
I do not remember ever hearing of mica caps going bad.? I do not think they age, or fail except by over-voltage stress.
?
But one hears about electrolytics (both Al and Ta) aging and/or going bad all the time.? It seems to be in their nature.? Chemical reactions, and all that.
?
Do you have a DC problem?? Might the circuit board need cleaning?
?
I would like to know the real story about the LM121 - why they discontinued it, and why someone had the gall to re-use the part number.? If the chip itself disappeared from all production in the late 1970s, it is possible nobody ever made its SPICE model.? I do remember that AppNote (AN-79), but at the time I was much less interested in the use of chopper stabilized amps, so I probably gave little attention to the original LM121 series.
?
Andy
?
?
|
True Andy. I meant to refer to the LM108 in the app note. The circuit I have is using the LM118 though. Still looking for that LM121 model.?
?
The issue is that the circuit I have is not bench performing the way one would expect. Thus I am wondering if it is possibly the mica caps that are use; though I have not heard of mica caps going bad as often as tantalum and electrolytics. I have swapped out with new LM121 and LM118 parts with not change in the circuit operation. Thus, I was looking for a model so I can play with some other values to see if I can "break" the circuit and get the simulation to behave somewhat as I see on the bench.
|
On Wed, Mar 12, 2025 at 02:59 PM, DerekK wrote:
... I found the LM118, as I can use the LTSpice LT118A model. Now for the front end.
Careful, there!? You have the wrong second-stage op-amp.? 118 ¡Ù 108.
?
The application note shows the LM108A as the op-amp that is coupled with the LM121 preamp.? It is not a LM118A.? Those two op-amps, LM108 and LM118, are very different from one another.
?
If I remember correctly, the LM108A was a somewhat slow, precision op-amp with very low input bias current and noise.? Whereas the LM118 was a screaming fast op-amp with less precision (higher offset voltage), more noise, and much higher input bias current.? I think it would not make a good substitute for the LM108A.
?
Linear Tech second-sourced the LM108 series, and LTspice comes with models for both LM108A and LM308.? (They might even be the same SPICE model as one another, but I didn't check that.)
?
Linear Tech also offered a LM118 substitute, but "improved" it somewhat, and changed its name from LM118 to LT118A.? Notice the changed prefix from LM to LT.? The name change means it is not a 1-for-1 replacement for the standard LM118.? The LT118A SPICE model that comes with LTspice would apply to their modified product (now obsolete), but maybe not to LM118 parts in general.? I do not know what is different about it.? Just be aware that it is not the same.? And it is definitely not the same as the LM108A.
?
Andy
?
|
On Wed, Mar 12, 2025 at 05:38 PM, John Waugaman wrote:
...? The LM321 is?not?the same device!
I think what you mean to say is that TODAY's LM321 is not the same device!
?
But originally the LM321 was the same as the original LM121.? Then at some later date, either NatSemi or Texas Instruments scrapped the old LM121/221/321 part number (and chip), and re-assigned the number to a completely different animal!? Today's LM321 is a conventional single-output high-gain op-amp designed for low-power - whereas yesterday's LM321 was a diff-in/diff-out preamplifier or in-amp, with much too little voltage gain to be called an op-amp.
?
I would not have thought it possible, but apparently it is.? Who would have thought it?
?
Andy
?
|
Chan model for saturable transformer LTSPICE simulation
#Transformer
Hi all, I'm working on a project that involves saturable transformer and when I try to model it in LTSpice, I can observe the saturation effects when there is sinusoidal voltage but I cannot really observe it when I supply pulsating DC voltage. Please find the LTSPICE files in here: /g/LTspice/files/Temp/SaturableTransformer.zipI went through all the information available on the internet and it helped me a lot. I am stuck with this since some time now. Can anyone please help me with it?
|
Mike, that is the correct datasheet. The LM121 was mentioned in the 1994 NSC Application Handbook. I uploaded that ap note to the TEMP files as well (LM121 AN-79.pdf).
|
All,
I just uploaded a LM121 data sheet to the temp folder.
Mike
|
On 3/12/25 4:38 PM, John Waugaman via groups.io wrote: The LM121 is shown as a preamplifier IC in the 1976 National Semiconductor Linear Data Book, but it's not listed in the 1980 edition. Strange, my yellowing copy of the 1980 edition has the LM121/221/321 on page 4-11. -- David Schultz "The cheeper the crook, the gaudier the patter." - Sam Spade
|
Indeed; many web sites offering its data sheet
do not in fact include LM121, only LM221 and LM321, which do not
have two outputs.
On 2025-03-12 21:38, John Waugaman via
groups.io wrote:
The
LM121 is shown as a preamplifier IC in the 1976 National
Semiconductor Linear Data Book, but it's not listed in the
1980 edition.? I will look to see if I still have the 1976
data book or any other data for this device.? The LM321 is?not?the
same device!
John
--
OOO - Own Opinions only
If something is true:
* as far as we know - it's science
*for certain - it's mathematics
*unquestionably - it's religion
|
The LM121 is shown as a preamplifier IC in the 1976 National Semiconductor Linear Data Book, but it's not listed in the 1980 edition.? I will look to see if I still have the 1976 data book or any other data for this device.? The LM321 is?not?the same device!
John
|
The old National Semi numbering system was LM1xx for Mil-temp devices, LM2xx for industrial temp, and LM3xx for commercial temp. Then the LMyx1 was for the single version, LMyx2 for dual and LMyx4 for the quad.?
?
The LM121 I am looking for is unique as it has a differential output and is paired with another opamp such as the LM118 to become a low-drift, high-gain set.
|
On Wed, Mar 12, 2025 at 11:59 AM, DerekK wrote:
I am trying to work out some old schematic and circuit issues and am looking for a model for the old LM121 opamp. Anyone have such a thing? I found the LM118, as I can use the LTSpice LT118A model. Now for the front end.
If you meant LM321, it is a single version of an LM324.
|
The LM121 has a differential output. Pinout is?
1: output 2
2: IN-
3: IN+
4: V-
5: balance
6: balance
7: V+
8: output 1
?
|
All the models do not show the proper LM121 differential output configuration. Andy, you are correct about the LM121 being special. The old National Semi Apps Handbook has the LM121 and LM118 used as a high-gain, low-drift amplifier set. Hence, the differential output of the LM121 is fed into the LM118 directly.
|
It looks like the LM121/LM321 was a very ordinary 1 MHz GBP op-amp, but maybe with somewhat lower operating power (lower supply current).? It is capable of +/-16 V supplies, but optimized for 5 V (+/-2.5 V) power.
?
T.I. has a product webpage for it: .? Near the bottom of that page, they include some PSpice and TINA-TI models? for it, which, oddly enough, are the SAME models as for the LM158/LM358/LM2904 garden-variety op-amps.
?
This suggests that you can substitute almost any generic 50-year-old op-amp for the part, including the bad-old 741 - at least for your SPICE simulations.
?
Andy
?
?
|
... I found the LM118, as I can use the LTSpice LT118A model. ...
As I say, it's been too many years.? But I do not remember that the LM121 was anything like the LM118.? The latter was optimized for high slew rate and bandwidth, unusually high in its day.
?
Andy
?
|
On Wed, Mar 12, 2025 at 02:59 PM, DerekK wrote:
I am trying to work out some old schematic and circuit issues and am looking for a model for the old LM121 opamp. Anyone have such a thing?
Remember that LM121, LM221, and LM321 are the same, just with different worst-case ratings.
?
There is something that claims to be an LM321 SPICE model, here:
It is a text file.? You can rename it if you feel like it.
?
There is a test schematic that uses it, here:
?
All you need is LTspice's built-in "opamp2" symbol, with the name changed to match that of the .SUBCKT model, and add a ".lib" or ".inc" command to include the model itself.? It's easy.
?
CAUTION:? That model is actually for Maxim's LMX321, which is a low-voltage version of the LM321.? I don't have a LM121/LM321 datasheet handy, but I suspect the original was not a low-voltage op-amp, like the LMX321 is.? So, caution is called for.
?
Turning now to the PSpice model at the previously-referenced T.I. webpage for their LM321LV - it is not an encrypted model, so chances are good-to-excellent that it works in LTspice.? Most PSpice models are SPICE, and LTspice understands SPICE and most PSpice models quite well.? Forget about all that Orcad stuff.? The .lib file is the SPICE model so it is the only one needed.? Once again, use the "opamp2" schematic symbol.
?
The TINA model there is not SPICE, so don't try that in LTspice.
?
Now I wonder whether an LM321LV is a suitable replacement for the LM121/LM321.? The "LV" in the part number suggests that it is not.? It is indeed a low-voltage op-amp, so it might not work on your old schematics.
?
Can you tell us, what made the LM121 unique?? It's been so long....
?
Andy
?
?
|
DerekK
John Woodgate
John Waugaman