Timeline for Do I really need the open loop output impedance of an op amp?
Current License: CC BY-SA 3.0
22 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 6, 2015 at 18:12 | comment | added | thexeno | @spehro The simulation shows a single pole at 1Mhz and amplitude is practically grounded to 0V after 10Mhz. I read the output of the phase from the sensing resistor. Next step: try to calculate this on the paper and mount that (as soon as I can gen put my hands on a signal generator). Parameters: Rs 0.33ohm and Vin 0.33Vac (0.165Vdc) with LM324+BUZ70 | |
| Jan 6, 2015 at 16:30 | comment | added | Spehro 'speff' Pefhany | It's not that bad- but a lot of times people find an amplifier oscillates or is only marginally stable because they didn't consider some side effect. Assuming R0 is negligible is such an assumption that can come back and bite you on the hindquarters. | |
| Jan 6, 2015 at 16:26 | comment | added | thexeno | @spehro your AD link says that you don't have to trust SPICE when considering Ro. Moreover, it's quite discouraging when says "the design procedure turns into a guessing game-and a prototyping nightmare". So, seems that this is a circuit that needs a lot of empirical design. | |
| Jan 5, 2015 at 16:07 | comment | added | thexeno | I will do that, but I will talk about that here: electronics.stackexchange.com/questions/146297/…, since I'm trying to split the problems. | |
| Jan 5, 2015 at 15:27 | comment | added | Spehro 'speff' Pefhany | @thexeno Why not just simulate your entire circuit? | |
| Jan 5, 2015 at 15:26 | comment | added | thexeno | Ok. Just to focus better the problem: to achieve a low precision in the equations (>20%) of the first and second pole frequencies, should I still make that sweep in SPICE and modelling the impedance to be put in the Ro variable, or a more general family assumption can be made as suggested by gsills? (note that I didn't yet read the others link, the answer may be there) | |
| Jan 5, 2015 at 15:19 | comment | added | Spehro 'speff' Pefhany | @thexeno Only because I chose a low frequency for the stimulus. I could do a sweep and get magnitude or magnitude and phase across a wide range. Of course you can't do this so simply in reality. | |
| Jan 5, 2015 at 15:16 | comment | added | thexeno | @Spehro But your method gives you the Rout at a low frequency only. I think I can use it as a starting point, anyway. (I still need to find some time to read a bit deeply the AD link and the stability unitygain linked in the other answer). | |
| Jan 5, 2015 at 0:34 | comment | added | Spehro 'speff' Pefhany | @RespawnedFluff Without any fancy reverse engineering, I simply connected an LM324 in PSPICE with offset applied so it was within 100mV or so of 0V (+/- 7.5V supplies) and nothing connected to the output. Applied a 2mA p-p 10Hz sine wave to the output and got 128.54mV swing at the output (about +18mV to -109mV), so output resistance is 64.3 ohms. Not unreasonable. | |
| Jan 5, 2015 at 0:10 | comment | added | user16497 | Yeah, I didn't see it at first, but something made me do a double take. What? | |
| Jan 4, 2015 at 23:52 | comment | added | Spehro 'speff' Pefhany | @gsills Wow, that could trip up a neophyte. I never noticed it was wrong, and I guess neither did the authors. | |
| Jan 4, 2015 at 23:18 | comment | added | user16497 | Looks like the schematic in the AD link for in-loop-compensation is wrong, since it shows positive feedback for the OpAmp. Otherwise good link. | |
| Jan 4, 2015 at 22:31 | comment | added | got trolled too much this week | The suggestion to look in the SPICE model works better for the more mundane NE5532/NE5534. TI gives only a bog standard Boyle macromodel with RO1 50 and RO2 25. So at DC the open-loop resistance is 75 ohm and at high frequency it is 50 ohm. | |
| Jan 4, 2015 at 19:47 | comment | added | Spehro 'speff' Pefhany | Well, if you're looking for 0.1% accuracy at a gain of 1 you only need about an open-loop gain of 1000, so it sounds like you're fine as even the worst op-amps generally have minimum guaranteed gains in the tens of thousands. Things get a bit fussier with high-Q active filters- you might need to multiply by the Q of the filter. | |
| Jan 4, 2015 at 19:44 | comment | added | thexeno | A bell sound in my head when I read "in comparison to the required accuracy". I know that the loopgain is quite near to Ad, because of the near to one voltage gain. I need to think to that, because all of your precious help is related to some exotic tecniques which are normally not used. My final goal is to find a correct frequency analysis with a 5%/10% of accuracy only (given by the components), nothing fancy at all. | |
| Jan 4, 2015 at 19:37 | comment | added | thexeno | Thanks for the link. I will take a look. Seems not a simple question to know the Ro of the opamp, isn't it? On my old notes of electronic, I've found the frequency compensation with capacitive loads. The high freq pole, which depends on the Ro, would be under 0dB with some RC compensation. But to quantify where it is, a literal Ro is present inside the equation. In practical exercises, we have always neglect the capacitance on loads: in those cases where otherwise Ro would be needed. | |
| Jan 4, 2015 at 19:37 | comment | added | got trolled too much this week | By the way, the LM124 model from TI appears to be a MPZ, not Boyle, given the output inductor it has. | |
| Jan 4, 2015 at 18:36 | comment | added | Spehro 'speff' Pefhany | I refer you to the linked AD application note where it advises against it. ;-) | |
| Jan 4, 2015 at 18:35 | comment | added | got trolled too much this week | I have actually looked at the LM124 SPICE model (from Nat Semi now TI) and compared it with the Boyle macromodel (paper), which uses two resistors (one is the AC output impedance at high frequency, and their sum gives the DC output impedance) but I'm having trouble applying that scheme to the actual LM124 SPICE model, which has 3 resistors (and an inductor) in its output stage. Since you mentioned reversing the SPICE model (evil grin), it might be illustrative if you could work out the LM124 as an example. | |
| Jan 4, 2015 at 17:13 | history | edited | Spehro 'speff' Pefhany | CC BY-SA 3.0 |
added 1165 characters in body
|
| Jan 4, 2015 at 16:17 | comment | added | thexeno | I made an EDIT, which seems to be more coherent with your suggestion to be not dependent on undeclared values. Check it out and let me say if it's right, if you want. | |
| Jan 4, 2015 at 14:30 | history | answered | Spehro 'speff' Pefhany | CC BY-SA 3.0 |