Post by thread:[EE]: Unused pins on a comparator

This reminds me of something I came accross recently on an LM339 comparator. I have a book that says to ground all unused pins of the comparator. I would think that with both input pins grounded that the output would be unstable. It seems to make more sense to me to ground one input and connect the output to VCC to get a stable output. Anyone not agree? Dave Dilatush wrote: {Quote hidden}>Rick Mann wrote... > >>What should I do with the unused pins on a dual CMOS op-amp (specifically, >>the National LM6482 rail-to-rail op amp)? If anything, is it enough to tie >>the inputs of the unused op amp together? Should they also be grounded? >> >>The datasheet had no recommendations, nor did the FAQ. >> > >You can tie one opamp input (doesn't matter which one) to one supply >rail, and the other opamp input to the other rail, and leave the output >open; or you can tie the output to the inverting input to make a >unity-gain follower, and then tie the non-inverting input to ground or >to either of the rails. > >Either way, the output of the opamp is determinate and stable- which is >what you want so the unused opamp doesn't interfere with anything. > >HTH... > >Dave > >-- >http://www.piclist.com#nomail Going offline? Don't AutoReply us! >email spam_OUTlistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body > > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

At 12:34 PM 3/2/02 -0800, you wrote: >This reminds me of something I came accross recently on an LM339 >comparator. I have a book that says to ground all unused pins of the >comparator. I would think that with both input pins grounded that the >output would be unstable. It seems to make more sense to me to ground >one input and connect the output to VCC to get a stable output. Anyone >not agree? I'd suggest following the original manufacturer's data-sheet recommendation, which is to ground all 3 pins of unused comparators (that includes the output). (In unusual circumstances, there may be some benefit in low-level reduced noise or current consumption with biasing it one way or the other, but I'd prefer to see ~100mV to do it, not Vcc). Connecting an open-collector output to Vcc is not a very good idea, but you probably didn't actually mean to write that! Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam@spam@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com 9/11 United we Stand -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Brian Kraut wrote... >This reminds me of something I came accross recently on an LM339 >comparator. I have a book that says to ground all unused pins of the >comparator. I would think that with both input pins grounded that the >output would be unstable. It seems to make more sense to me to ground >one input and connect the output to VCC to get a stable output. Anyone >not agree? The idea of grounding all unused pins--both inputs AND the output--of an unused LM339 section works because the LM339 output is open-collector; it can only pull down. You wouldn't want to do this with any device having an active-high output (such as an opamp), but doing it with an LM339 comparator is OK. This applies only to an LM339 whose negative supply terminal is grounded; obviously, if it is tied to a negative voltage, the unused output should be tied there, not to ground. For the LM339, connecting one input to ground and the other input to VCC would be pointless because the LM339's input common-mode range does not include VCC; and when either of the inputs is outside the input CMR, the output is indeterminate. Dave -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

O.K. Obviously the open collector tied to ground will be a determined output, but I guess it can keep oscillating between open collector and ground internally. I guess that must be O.K. though. I thought that maybee it was a typeo in my book. This is probably the only practical soultion since grounding one input and having one at some less than Vcc voltage divider would be impractical. The output will probably stay one way or the other with both inputs grounded from tollerences in the part anyway. Dave Dilatush wrote: {Quote hidden}>Brian Kraut wrote... > >>This reminds me of something I came accross recently on an LM339 >>comparator. I have a book that says to ground all unused pins of the >>comparator. I would think that with both input pins grounded that the >>output would be unstable. It seems to make more sense to me to ground >>one input and connect the output to VCC to get a stable output. Anyone >>not agree? >> > >The idea of grounding all unused pins--both inputs AND the output--of an >unused LM339 section works because the LM339 output is open-collector; >it can only pull down. You wouldn't want to do this with any device >having an active-high output (such as an opamp), but doing it with an >LM339 comparator is OK. This applies only to an LM339 whose negative >supply terminal is grounded; obviously, if it is tied to a negative >voltage, the unused output should be tied there, not to ground. > >For the LM339, connecting one input to ground and the other input to VCC >would be pointless because the LM339's input common-mode range does not >include VCC; and when either of the inputs is outside the input CMR, the >output is indeterminate. > >Dave > >-- >http://www.piclist.com hint: PICList Posts must start with ONE topic: >[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Brian Kraut wrote... >O.K. Obviously the open collector tied to ground will be a determined >output, but I guess it can keep oscillating between open collector and >ground internally. I think that's what's called a "distinction without a difference". The bottom line, when dealing with unused opamp sections or comparator sections, is twofold: 1. Keep both inputs within their specified common-mode range. 2. Keep the output from flopping around. Whatever satisfies both of these requirements, works. Dave -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

At 05:43 PM 3/3/02 +0000, you wrote: >For the LM339, connecting one input to ground and the other input to VCC >would be pointless because the LM339's input common-mode range does not >include VCC; and when either of the inputs is outside the input CMR, the >output is indeterminate. Hmmm.. "indeterminate" or "unspecified"? ;-) From the internal schematic, it looks like tying both pins to a voltage >> Vcc (but < 36V) will lead to a predictable result (output low and and some current saving). There's a range around Vcc where the current sources are shutting down that won't be so predictable if both inputs are in that same range, but if one pin is within the common-mode range, the results from the other pin being -0.3 < Vin < 36V should not be different from what you'd expect, regardless of supply voltage. (BTW, their assumption that ground = negative supply was corrected on later data sheets, I guess some people didn't realize the reasoning) Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" speffKILLspaminterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com 9/11 United we Stand -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Works for me. Thanks. Dave Dilatush wrote: {Quote hidden}>Brian Kraut wrote... > >>O.K. Obviously the open collector tied to ground will be a determined >>output, but I guess it can keep oscillating between open collector and >>ground internally. >> > >I think that's what's called a "distinction without a difference". > >The bottom line, when dealing with unused opamp sections or comparator >sections, is twofold: > >1. Keep both inputs within their specified common-mode range. > >2. Keep the output from flopping around. > >Whatever satisfies both of these requirements, works. > >Dave > >-- >http://www.piclist.com hint: PICList Posts must start with ONE topic: >[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> This reminds me of something I came accross recently on an LM339 > comparator. I have a book that says to ground all unused pins of the > comparator. I would think that with both input pins grounded that the > output would be unstable. It seems to make more sense to me to ground > one input and connect the output to VCC to get a stable output. Anyone > not agree? I don't like tying both input pins together because, as you say, it could be unstable. You are counting on the offset voltage to drive the output one way. However, I would definitely NOT tie one to power and the other to ground. Depending on the input stage, that could draw excessive current or even destroy the chip. This is less likely to be a problem on comparators because these are usually designed for a large differential mode input. ******************************************************************** Olin Lathrop, embedded systems consultant in Littleton Massachusetts (978) 742-9014, .....olinKILLspam.....embedinc.com, http://www.embedinc.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads