| T O P I C R E V I E W |
| shadders |
Posted - 04 Apr 2019 : 17:14:26
Hi,
I am using the LM319 in a design, and the mapping between the Schematic symbol and PCB symbol are incorrect. I believe the LM319 is in the prolib library, and versions i checked were LM319AM, LM319M and LM319N.
The device is a dual op-amp with open collector outputs.
The error is that the schematic has the connections for op-amp1 and op-amp2 on the left and right side respectively, of the schematic symbol. The mapping has the output of op-amp1 on the op-amp2 pin, and vice versa. (pins 7 and 12).
Regards,
Shadders. |
| 9 L A T E S T R E P L I E S (Newest First) |
| Iain Wilkie |
Posted - 05 Apr 2019 : 20:01:31
Third party libraries sound great in principle but their not really.
I could probably count on one hand when I’ve used a library component without modifying it .... and generally speaking I always tend to generate my own, then I know it’s right and the way I want it.
Schematic symbols are the worst ! They tend to be the rectangular box type mimicking the package pinning, nine times out of ten this is not how you want it. Logical grouping of signals makes for better schematics for complex components and simpler components such as transistors, fets,triacs, opamps etc, etc, are much more meaningful in their true symbolic form than simple boxes.
As for the footprints well they tend be generalised and even though they may be functionally correct, a quick visit to the component manufacturers datasheet can uncover slight but important dimensional differences.
So I would be careful when deciding to use a third party component, check it very carefully and edit it accordingly if it’s not quite the way you want it. Alternatively simply create your own component, then you know it’s correct and the way you want it.
Iain |
| shadders |
Posted - 05 Apr 2019 : 11:09:47
Hi,
Yes - i made the assumption that the schematic symbol was logical, not a direct copy of the pin outs of the device.
Regards,
Shadders. |
| Iain Wilkie |
Posted - 05 Apr 2019 : 11:01:03
quote:
If Number One are not interested in these type of issues, then ok.
To be fair ..... there is no error ..... if fact the part appears the same in other libraries eg Symacsys.
Iain
|
| edrees |
Posted - 05 Apr 2019 : 10:33:28
quote:
Having the output of the left hand side inputs associated with the right hand side inputs, is illogical.
Maybe, but its the way National Semi designed the chip. No.1 have presented the component exactly as per the data sheet so I don't see it as a No.1 issue. |
| shadders |
Posted - 05 Apr 2019 : 10:21:25
Hi,
I agree now, never trust the third party library.
It depends on how you look at it. The symbol names are not as accurate as they could be, but in their current format, for the left hand side inputs, output is on the right hand side, and vice versa.
Using the device, i would assume the OUT on the left hand side of the symbol is the output for the left hand side inputs. This is logical. Having the output of the left hand side inputs associated with the right hand side inputs, is illogical.
The entire point of being able to map pins on the schematic symbol, to the package/footprint in the component editor, allows a logical symbol to be created in the schematic
If Number One are not interested in these type of issues, then ok.
Regards,
Shadders. |
| edrees |
Posted - 05 Apr 2019 : 09:33:58
The Prolib LM319 pinout (Schematic & PCB) is correctly mapped, but the Inputs/Outputs terminal names could be more specific, i.e. In1+, In1-, and Out1. You can onbiously edit these terminal names in the Component Editor if you so wish.
However, I see that the original Nat Semi Data sheet doesn't really highlight which output is which anyway! |
| Iain Wilkie |
Posted - 05 Apr 2019 : 09:31:21
Shadders ..... The problem is the symbol ... this one is simply a basic representation of the actual chip, there is no error, its just not a good symbol. I tend not to use symbols like this ... especially for opamps as the layout connections are not as informative as the usual triangular opamp symbol which makes reading a schematic much easier. In such circumstances I tend to design my own.
Regarding checking every component from any library before use .... well your missing a fundamental check ! .... you MUST always check, unless you created the part yourself, you cannot be 100% certain a pre-made part is correct. Indeed over the years I have come across many library parts with errors, either in the connection info or more often in the footprint, I don't trust any imported component.
Iain
|
| shadders |
Posted - 04 Apr 2019 : 22:18:10
Hi Iain,
The problem is, you automatically assume that the OUT on the left is based in the inputs In- and In+ on the left. The Out on the left is based on the inputs from the right of the schematic symbol.
Why would you check every component in the library with the datasheet before use ? Isn't that the whole point of an existing library - no need to check.
It is not a problem if you are using both opamps to pull down on the same common output. If you are using one opamp, then this is a problem.
Regards,
Shadders. |
| Iain Wilkie |
Posted - 04 Apr 2019 : 19:58:38
When I pull up the component up (LM319M) there is no distinction between the two opamps so not a problem as you would use the connections referring to the datasheet.
Iain
|
|
|
