The Jensen 990 will phase reverse in simulation even when you remove the input protection diodes. The JFET-input TL072 phase reverses and yet it has no input protection diodes. The PNP-input LM358 & LM324 phase reverse and they have no input protection diodes.
It's a quick protoboard exercise to connect up an actual opamp chip as a unity gain follower, then drive its input with an almost-rail-to-rail sinewave while watching the output. You could lash together the setup and test four different opamps, within 60 minutes. If it was important to you.
It's a quick protoboard exercise to connect up an actual opamp chip as a unity gain follower, then drive its input with an almost-rail-to-rail sinewave while watching the output. You could lash together the setup and test four different opamps, within 60 minutes. If it was important to you.
The phase reversal issue was mostly a tempest in a teapot. There are op-amps that don't get even 1V from the rail or have a heavily compromised input common mode range that will also burn you if you don't read the datasheet.
The phase reversal issue came up most strongly in things like PID controllers where the phase of the feedback gets inverted (positive feedback ->latch-up). Phase reversal in a non-feedback situation (that is a forward gain path only) would just be, when you clip it's REALLY distorted.
The phase reversal issue came up most strongly in things like PID controllers where the phase of the feedback gets inverted (positive feedback ->latch-up). Phase reversal in a non-feedback situation (that is a forward gain path only) would just be, when you clip it's REALLY distorted.
Last edited:
Interesting thread on newer low noise devices:
http://www.diyaudio.com/forums/solid-state/243432-where-have-all-low-noise-transistors-gone.html
And this, too bad the duals are NLA:
http://www.diyaudio.com/forums/solid-state/92901-has-anyone-used-these-gems.html
winhill2 wrote:
"There are lower-noise transistors, lower e_n anyway, with r_bb' down to the under 2-ohm region, see Table 8.1 in AoE 3rd edition for full details, but they're not specified as low-noise transistors. They're low-saturation-voltage designs made by Zetex / Diodes, Inc. For example the ztx618 and ztx718 (fmmt618 and 718 for sot-23 packages) with r_bb' of 9.3 and 7.3 ohms, and the ztx851 and ztx951 with r_bb' of 1.67 and 1.24 ohms. We use the latter (multiple pnp transistors in parallel) to make a record-breaking low-noise amplifier, only 70 pV/rt-Hz. That's low enough to be a ribbon-mic pre-amp, without the usual step-up transformer!"
Looks like I'm going to have to buy AoE 3rd edition!
http://www.diyaudio.com/forums/solid-state/243432-where-have-all-low-noise-transistors-gone.html
And this, too bad the duals are NLA:
http://www.diyaudio.com/forums/solid-state/92901-has-anyone-used-these-gems.html
winhill2 wrote:
"There are lower-noise transistors, lower e_n anyway, with r_bb' down to the under 2-ohm region, see Table 8.1 in AoE 3rd edition for full details, but they're not specified as low-noise transistors. They're low-saturation-voltage designs made by Zetex / Diodes, Inc. For example the ztx618 and ztx718 (fmmt618 and 718 for sot-23 packages) with r_bb' of 9.3 and 7.3 ohms, and the ztx851 and ztx951 with r_bb' of 1.67 and 1.24 ohms. We use the latter (multiple pnp transistors in parallel) to make a record-breaking low-noise amplifier, only 70 pV/rt-Hz. That's low enough to be a ribbon-mic pre-amp, without the usual step-up transformer!"
Looks like I'm going to have to buy AoE 3rd edition!
Last edited:
Almost ashamed to show my head in here (I know people like Winfield Hill hang around and ... you know fanboi and all that).
Anyway, admin kindly re-opened this thread for me and I've been studying the paper quite a bit - I don't know how to work in the time domain so Deane's calculus is a bit over my head - so far it's out of Earth's gravity.
Most of the rest I understand, but one of the later revisions added a couple of IN914s in series with the inverting collector on the differential pair and I just can't understand what function they perform.
I'd also like to drop (gasp) that neat little inductor trick on the emitters as I'm designing for different transistors and I assume (you know what assume did) it's something that can be omitted in as can the 30R resistors since that's all well above the frequencies of interest and I can presumably put a correction in the feedback loop by paralleling a capacitor on on the feedback resistor - per an integrated amp.
Although I suspect this will work quite well with intrinsically matched MMDT3904s, I'm leaving space in the board to replace them with a MAT12 when budget allows. As luck would have it, the MAT12 seems very similar operationally to the LM394 and runs at a 1mA per transistor for best performance, not that far away leaving it with an Re of 25R I believe. Deane ran each transistor at 2mA for an Re of 16R. Not a huge discrepancy but it will affect the other calculations.
The real annoyance with the MMDT3904s is their ideal low-noise collector current at just 100uA for an Re of 250R if I'm doing my sums right and that's gonna play havoc with all those poles and zeros.
I've pushed LTSpice about as far as I dare but I know it has limitations that can bit you on the fundament (I've seen a few when I cocked up by forgetting to include a ground or a current source goes outside of compliance but I trust the maths, even if I'm not terribly good at it. I don't need a worked example (would just go over my head anyway) but even a pointer to where to get the formulae would be really handy. As I've said, I'm more comfortably in the frequency domain so those would be my preferred equations. The only thing I know about Laplace Transforms is they are called Laplace transforms. I tried to follow a couple of talks but my brain is a bit passed it's best before date.
Anyway, admin kindly re-opened this thread for me and I've been studying the paper quite a bit - I don't know how to work in the time domain so Deane's calculus is a bit over my head - so far it's out of Earth's gravity.
Most of the rest I understand, but one of the later revisions added a couple of IN914s in series with the inverting collector on the differential pair and I just can't understand what function they perform.
I'd also like to drop (gasp) that neat little inductor trick on the emitters as I'm designing for different transistors and I assume (you know what assume did) it's something that can be omitted in as can the 30R resistors since that's all well above the frequencies of interest and I can presumably put a correction in the feedback loop by paralleling a capacitor on on the feedback resistor - per an integrated amp.
Although I suspect this will work quite well with intrinsically matched MMDT3904s, I'm leaving space in the board to replace them with a MAT12 when budget allows. As luck would have it, the MAT12 seems very similar operationally to the LM394 and runs at a 1mA per transistor for best performance, not that far away leaving it with an Re of 25R I believe. Deane ran each transistor at 2mA for an Re of 16R. Not a huge discrepancy but it will affect the other calculations.
The real annoyance with the MMDT3904s is their ideal low-noise collector current at just 100uA for an Re of 250R if I'm doing my sums right and that's gonna play havoc with all those poles and zeros.
I've pushed LTSpice about as far as I dare but I know it has limitations that can bit you on the fundament (I've seen a few when I cocked up by forgetting to include a ground or a current source goes outside of compliance but I trust the maths, even if I'm not terribly good at it. I don't need a worked example (would just go over my head anyway) but even a pointer to where to get the formulae would be really handy. As I've said, I'm more comfortably in the frequency domain so those would be my preferred equations. The only thing I know about Laplace Transforms is they are called Laplace transforms. I tried to follow a couple of talks but my brain is a bit passed it's best before date.
I suggest that you start a new thread with a link to this thread since you're not really building a
JE990 and it is off topic here. I like to help people but don't have a lot of time for this, perhaps
others will help you.
Tell us what is your application (MM, MC, Mic pre, or?) for this amp in the new thread.
JE990 and it is off topic here. I like to help people but don't have a lot of time for this, perhaps
others will help you.
Tell us what is your application (MM, MC, Mic pre, or?) for this amp in the new thread.