Driving an OP AMP input with no VCC

When you have the potential for input to exceed VCC or ground it is good practice to just clamp the signal at the input pin with diodes to carry the current up/down to VCC or ground. The diodes are reversed biased during normal ops.

Good practice for long wire runs to inputs to keep induced voltages out and in industrial and automotive environments to clamp EMI.

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Depends on the OpAmp. Read the maximum input voltage spec. If it defines maximum input as a relationship with VCC, then you may have problems.

Be more precise in you question... OpAmp type, etc.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

I am using and MCP6044 running at 5V and I have the input clamped via BAT54S diodes to a 3.3V rail and ground. The problem is that my board may loose power while an input remains. I have a TVS on the 3.3V rail to prevent it going above 3.3V but this would mean that an input would be clamped at 3.6V while the OP AMP's Vcc would be 0V.

Best regards

AJ

I am unable to view a data sheet from Microchip's site. Firefox just hangs and sucks 100% CPU. Sorry.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Thanks anyway mate, I am not committed to the MCP6044, can you recommend one that would be suitable for my application or tell me what I should be looking for?

Best regards

AJ

What is your application? If you are looking to use an op-amp as a follower, you could look at the OPA333 from Texas Instruments. Not that fast but really good spec's in terms of low power, low voltage, and offset (rail to rail too!).

With any op-amp (just about anyway) you need to pay attention to the data sheets absolute maximum ratings. Usually there will be a maximum rating on the input voltage;i.e. Vcc + 0.3V using the OPA333 as an example. If you go beyond this then you need to heed the maximum input current rating. Again using the OPA333 as an example, it is listed as

+/-10mA and it is explicitly stated in the datasheet as well. A simple series resistor can be used to do this. You can also use clamps on the inputs too. As far as I am aware, this limited input current is continuous and would not damage the part and I would go as far as to say that it would not cause worry in terms of degradation or reliability either. The real fun is what happens when you are using a dual or a quad and one channel experiences an electrical overstress event while the others are driven normally. In your case if you are not going to have power connected, the point is moot; however, through my discussions with several colleagues and looking at test data it appears that you can treat each channel as a separate block within the IC. It is true that they will all ultimately share the same power rail and that there are many paths for current to flow during an EOS event. I will have to look at the test again to see how it was set up if you are interested.

All inputs and outputs................... VSS ?0.3 V to VDD +0.3 V

I think you will find most, if not all OpAmps have this restriction, so you need some way to protect it.

Post your schematic on a.b.s.e and everyone here will take a look.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |

Jim Thompson a écrit :

So it's time for some update. Have a look at "over the top" opamps or comparators like LT1716 or LT1782... Bias current is going high over the supply rail, but it is OK for some applications.

--
Thanks,
Fred.

[snip]

It just occurred to me that the LM324 can take inputs > VCC.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |

"Fred Bartoli" wrote in message news:4551f17a$0$3869$ snipped-for-privacy@news.free.fr...

Here is a link to the schematic to make things a bit clearer.... http://61.9.815.52/OP_IN.JPG . The output connects to a PIC analogue input and the idea is to be able to detect a high, low or float condition. The MCP6044 has VSS ?0.3 V to VDD +0.3 V so maybe I will be ok with this configuration. I have done a bit of testing and I haven't blown an OP yet.

I have seen these "over the top" OP's, didn't think of them, thanks for the suggestion.

Best regards,

AJ

Thanks for your reply mate, the MCP6044 does have vcc + 0.3V and I am connecting to the input via 10K so with any luck I will be ok, I haven't managed to blow one yet.

^^^ ? ?

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |

Sorry, wrong address, I tend to do these things at 3am. Try this one http://61.9.185.52/OP_IN.JPG

Regards

AJ

AJ a écrit :

Buffering with an opamp is unnecessary for a simple detection. And with your 30V input, you still have the same pb, whatever the 3.3V supply status. And thanks to the high leakage of schottkies, a floating input could well read a false high or low at high temperatures, depending luck.

That one will clamp your input voltage and provides an output within the supply rails. If you're really cheap you can get rid of the 2 leftmost 100K resistors. Use any NPN-PNP signal transistors.

1M VCC ___ .----+----|___|-. | | | .-. | | 100K | | \\| | ___ | | |-+------+-|___|-< 0-30V '-'

"Fred Bartoli" wrote in message news:4552e171$0$25698$ snipped-for-privacy@news.free.fr...

Thanks a lot for your input Fred, very helpful.

Best regards

AJ