Shunt reg: can I change the opamp?

[mark_titano]

Hi guys,
I've got some questions reguarding this pws shunt regulator:

http://www.euronet.nl/~aespreng/Shuntregv11d/Manual.pdf

In the manual the suggested op amp is the opa655P and the author say that is not possible to substitute it with others op amp due to the special dual power pins.
But that op amp is discontinued, replaced by the opa656.

http://focus.ti.com/docs/prod/folders/print/opa655.html

http://focus.ti.com/docs/prod/folders/print/opa656.html

Now, TI says that the 656 is "an EXACT EQUIVALENT in functionality and parametrics to the compared device (655)" so can I use it in place of the 655 without modifications? The 656 has not 4 pws pins but two of them are not conncted so probably there isn't any problem chainging the 655 with the 656. I'm right?

Thank you

Mark

 

[jan.didden]

You should be OK. The reason that they discontinued the extra power pins is probably cost vs benefit. Extra powerpins can help to limit inductance of the connection wires which might be important in transimpedance applications. In your application, make sure you have the decoupling caps as close as possible to the power pins, and/or use an SMD chip and caps.

Jan Didden

 

[mark_titano]

Thank you Jan,

the layout is exactly the same you can see in the manual. I've found the gerber files ( http://www.euronet.nl/~aespreng/Shuntregv11d/ ) and I've made the boards. Then I started to look for the components and I found out that TI discontinued the OPA655.

The smd version of the 656 is supply in sample by TI so I can easily try to test it.
I've got some smd to dil adapter and maybe I can use them with a ceramic cap directly on the adapter.
I don't think this will be necessary, the board layout is very tight.

Mark

 

[jan.didden]

[snip]I've got some smd to dil adapter and maybe I can use them with a ceramic cap directly on the adapter.
I don't think this will be necessary, the board layout is very tight.

Mark

Yeah you should try that. In applications with very high feedback even a fraction of an inch track or adapter pin length can give rise to instabilities, but try this and see if it is stable. If you'r out of luck, one thing to try is a very small capacitor (few tens of pF) across the series resistor or ccs feeding the shunt. That will kill the hf gain and tame the loop.

Jan Didden

 

[mark_titano]

Thank you very much for the advice

Mark