When using a TL082 opamp as a simple voltage follower with a single supply, I noticed that, if I lower the input voltage near ground (about 1V and lower), the output suddenly slams to Vcc. Strange and dangerous behavior. I was assuming the output would, at least, stay as low as it is possible for the output stage configuration.
What kind of general purpose with similar slew rate and input impedance would be available without this drawback? It must be able to buffer signals with sinus, triangle and square wave forms up to about 50KHz with 2Vp-p.
You have found the phase reversal issue with the TL081/2/4 series. I didn't find 'phase reversal' in the TI data sheet but Analog Devices mentions it for the AD712 which is similar. Try a 5532.
Have you checked if the non symmetrical slew rate and cross over distortion of the LM358 can handle 50kHz square waves at 2 Vp-p with any reasonable distortion?
Maybe ac-coupling into the TL082 input biased at V+/2, you could still keep the input impedance fairly high and a relatively small capacitor required for reasonably low frequency range. Another method would be resistor divider between signal input and V+. Say RL goes from OA input to signal input and Ru goes from same OA input to V+. Then the quiescent OA input voltage is VDC=(VS*RU+(V+)*RL)/(RU+RL) where VS is signal DC bias. The signal ac-component attenuation is compensated for by adding gain to the configuration with feedback divider to (-) input.
dangerous behavior. I was assuming the output would, at
TL072 does that, too. I use ne5532 whenever I need to avoid this behaviour.
One question for everyone, does SMD version of ne5532 also come in the usually narrow so8 case? Can't seem to find it anywhere.
--
Siol
------------------------------------------------
Rather than a heartless beep
Or a rude error message,
See these simple words: "File not found."
Working with 9V supply. The signal is normally biased at 4.5V but can go lower by design to shift the DC level at output. I just checked with a LM358 (got one in my stock :) According to datasheet, slew rate is about 0.2V/uS. A 50KHz cycle is only
20uS so I feared it was too slow and... it is. The wave form looks more like triangular. May be I must stick with FET or MOSFET input to get a reasonnably higher slew rate (5V/uS min) and low input bias (
have you tested the OP-amp by tying the input to a common via a R to see if it does this? if memory serves i think this is a cmos input op-amp? is it possible your driving it with a bipolar type input and your dropping below the the cut off point thus maybe causing a floating input which to me would swing high ? or maybe the input of the op-amp has this in it internally ? just a thought..
--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5
Is there a cheap rail-to-rail dual opamp that'll do 15V min and be available in so8?
Ups, some googling just turned up tlc272, this might do it.
--
Siol
------------------------------------------------
Rather than a heartless beep
Or a rude error message,
See these simple words: "File not found."
I think it only does phase reversal when input goes close to positive rail, which is in my case a lot easier to circumvent than the other extreem.
Actually, TLC272 input should allow voltage lower than negative supply rail, if I understand this datasheet correctly. It looks like something you could use, its cheap too @0.28 euro / piece.
--
Siol
------------------------------------------------
Rather than a heartless beep
Or a rude error message,
See these simple words: "File not found."
i was just looking at the data sheet, input common mode Vcm is +15.. -12 i don't see where you should have a problem? how ever, i do see a problem if there is not input R shunt? i mean, if the source of voltage goes down to 1.0 or lower when this takes place? this tells me that your reference input may also be going into into Hi Z ?, the 082 are Fet inputs, i can see where it could take off on a fast skew to the bottom possibly dropping into Hi Z and thus having a brief reverse of Capacitance discharge ?
i may have this incorrectly placed in my head but its just a thought and a very basic description on my part!
P.S. i have used 082's in common mode and haven't had this problem how ever, the circuits normally insure that there is a input shunt load on the Op-amp. how ever, i have seen cases where diodes or devices like it (something like an emitter flower transistor) feeding the input drop below the cut off point of the PN junction and thus bring the input to the cmos device to a Hi-z state very quickly. undesirable effect and not healthy for the device.
--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.