|
|
Post by thetragichero on Jun 15, 2022 16:08:50 GMT -5
picking up a schecter guitar this weekend for about the price i can flip the aftermarket pickups for. so, basically free guitar. i have an active bridge pickup for it (Duncan designed version of the mick thomson blackout), might install a sustainer in the neck because i don't have a neck pickup for it and because I've been wanting to do it for quite awhile (just need to find where my fargen coil wire got to in the move) one thing I'm not sure on: what to use for the tone pot hole. not a big fan of the standard cap abs pot tone control in my metal tele (even after using appropriate values for active pickups) so I'm soliciting options. i feel frets and MattB especially might be helpful one thought i had was buffer into a modified big muff tone control (values tweaked for that mids) into recovery gain stage into passive volume control. would the buffer be needed since I'm not using passive pickups? other idea is a little more extreme (but could work well with a metal guitar): basically distilling the gyrator eq stages from the boss heavy metal pedal to one knob so bass/treble boost and cut (I've got some nice s/w taper center detente pots that would work well here) |
|
|
|
Post by frets on Jun 15, 2022 19:19:04 GMT -5
Trag, What about a nice onboard EQ circuit?  I think this could be distilled into an onboard - maybe. |
|
|
|
Post by thetragichero on Jun 15, 2022 19:42:34 GMT -5
really just looking for something using one knob since that's what it's already drilled for
|
|
|
|
Post by MattB on Jun 16, 2022 9:45:40 GMT -5
one thought i had was buffer into a modified big muff tone control (values tweaked for that mids) into recovery gain stage into passive volume control. would the buffer be needed since I'm not using passive pickups? other idea is a little more extreme (but could work well with a metal guitar): basically distilling the gyrator eq stages from the boss heavy metal pedal to one knob so bass/treble boost and cut (I've got some nice s/w taper center detente pots that would work well here) You can (sort of) do both.  This is two Big Muff tone controls on a dual-gang pot, centred at different frequencies. SW1 flips one of the gangs backwards. With SW1 up, you get a mid scoop/mid boost. With SW1 down you get something similar to a regular Big Muff tone control, but with a mid boost instead of a cut.  These graphs were made with a 0dB input, volume at 100% and tone stepped in 20% increments. Increasing R9 and R16 to 47k would give you about 5dB more gain.
This is still very much a work in progress. I haven't spent a lot of time trying to fine-tune it. I did add C3, R5 and R12 to try and balance things out a little better.
It was going to be a two knob tone control for a fuzz pedal, but I haven't had time to work on it. I think there's plenty of room for improvement, but it could be a lot of work to get something worthwhile. Here's a simpler option. This is a fairly basic treble boost/cut. It uses a combination hi-pass/low-pass filter, with the idea being that you can increase brightness without adding too much fizz.
  And here's the same circuit, with C5 reduced to 1n. The high end roll off is less extreme, and the range of the pot is shifted upwards. You have a bit more boost and a lot less cut available.
 |
|
|
|
Post by thetragichero on Jun 17, 2022 18:51:03 GMT -5
MattB did you do that schematic in some sort of spice program? just downloaded kicad to figure out how to make a pcb on the little cnc (i have a bunch of copper clad g10 board from tayda) and figure not having to put in the whole thing would be easier for me (lazy)
|
|
|
|
Post by MattB on Jun 18, 2022 11:03:18 GMT -5
MattB did you do that schematic in some sort of spice program? just downloaded kicad to figure out how to make a pcb on the little cnc (i have a bunch of copper clad g10 board from tayda) and figure not having to put in the whole thing would be easier for me (lazy) Yes, I built it in LTSpice. Here is a copy of the simulation file:
Version 4
SHEET 1 2008 680
WIRE -1360 -624 -1584 -624
WIRE -1136 -624 -1360 -624
WIRE -1136 -592 -1136 -624
WIRE -1584 -528 -1584 -624
WIRE -1136 -496 -1136 -512
WIRE -1136 -496 -1232 -496
WIRE -1008 -496 -1136 -496
WIRE -1232 -480 -1232 -496
WIRE -1824 -464 -1920 -464
WIRE -1360 -448 -1360 -624
WIRE -1136 -448 -1136 -496
WIRE -1920 -400 -1920 -464
WIRE -1584 -368 -1584 -464
WIRE -1360 -368 -1584 -368
WIRE -1232 -368 -1232 -416
WIRE -1232 -368 -1360 -368
WIRE -1136 -368 -1232 -368
WIRE -1824 -224 -1824 -464
WIRE -2032 -160 -2192 -160
WIRE -1920 -160 -1920 -320
WIRE -1920 -160 -1968 -160
WIRE -1872 -160 -1920 -160
WIRE 16 -144 -96 -144
WIRE 272 -144 96 -144
WIRE 672 -144 272 -144
WIRE 272 -128 272 -144
WIRE 672 -128 672 -144
WIRE -1920 -112 -1920 -160
WIRE -1824 -112 -1824 -128
WIRE -1728 -112 -1824 -112
WIRE -1584 -112 -1664 -112
WIRE -1488 -112 -1504 -112
WIRE -1424 -112 -1488 -112
WIRE -1024 -112 -1424 -112
WIRE -96 -112 -96 -144
WIRE 848 -112 848 -192
WIRE 944 -112 848 -112
WIRE 1088 -112 1024 -112
WIRE -1424 -96 -1424 -112
WIRE -1024 -96 -1024 -112
WIRE -432 -48 -432 -128
WIRE -336 -48 -432 -48
WIRE -192 -48 -256 -48
WIRE 272 -32 272 -64
WIRE 432 -32 272 -32
WIRE 544 -32 512 -32
WIRE 672 -32 672 -48
WIRE 672 -32 624 -32
WIRE -1424 0 -1424 -32
WIRE -1392 0 -1424 0
WIRE -1024 0 -1024 -16
WIRE -1024 0 -1232 0
WIRE -1488 16 -1488 -112
WIRE -1920 32 -1920 -32
WIRE -1872 32 -1920 32
WIRE -1824 32 -1824 -32
WIRE -1824 32 -1872 32
WIRE 944 48 944 -16
WIRE -2192 64 -2192 -160
WIRE 848 64 848 -112
WIRE 912 64 848 64
WIRE 1088 80 1088 -112
WIRE 1088 80 976 80
WIRE 1104 80 1088 80
WIRE 1248 80 1168 80
WIRE 1424 80 1328 80
WIRE 480 96 480 16
WIRE 720 96 480 96
WIRE 816 96 784 96
WIRE 912 96 816 96
WIRE -1280 112 -1280 48
WIRE -784 112 -1280 112
WIRE -336 112 -336 48
WIRE -432 128 -432 -48
WIRE -368 128 -432 128
WIRE -192 144 -192 -48
WIRE -192 144 -304 144
WIRE -96 144 -96 -48
WIRE -96 144 -192 144
WIRE -560 160 -736 160
WIRE -464 160 -496 160
WIRE -368 160 -464 160
WIRE 1424 176 1424 80
WIRE 272 208 272 -32
WIRE 672 208 672 -32
WIRE -464 240 -464 160
WIRE -1424 256 -1424 0
WIRE -1024 256 -1024 0
WIRE -2192 352 -2192 144
WIRE -1872 352 -1872 32
WIRE -1872 352 -2192 352
WIRE -1488 352 -1488 80
WIRE -1488 352 -1872 352
WIRE -1424 352 -1424 336
WIRE -1424 352 -1488 352
WIRE -1024 352 -1024 320
WIRE -1024 352 -1424 352
WIRE -784 352 -784 192
WIRE -784 352 -1024 352
WIRE -336 352 -336 176
WIRE -336 352 -784 352
WIRE 64 352 -336 352
WIRE 272 352 272 288
WIRE 272 352 64 352
WIRE 672 352 672 272
WIRE 672 352 272 352
WIRE 944 352 944 112
WIRE 944 352 672 352
WIRE 1424 352 1424 256
WIRE 1424 352 944 352
WIRE -464 544 -464 320
FLAG 64 352 0
FLAG -1360 -368 0
FLAG -1360 -624 VCC
IOPIN -1360 -624 Out
FLAG -336 48 VCC
IOPIN -336 48 In
FLAG -432 -208 VREF
IOPIN -432 -208 In
FLAG 944 -16 VCC
IOPIN 944 -16 In
FLAG -1824 -464 VCC
IOPIN -1824 -464 In
FLAG -464 544 VREF
IOPIN -464 544 In
FLAG 816 176 VREF
IOPIN 816 176 In
FLAG 848 -272 VREF
IOPIN 848 -272 In
FLAG 1424 80 OUT
IOPIN 1424 80 Out
FLAG -1008 -496 VREF
IOPIN -1008 -496 Out
SYMBOL cap -1440 -96 R0
SYMATTR InstName C1
SYMATTR Value 3.9n
SYMBOL cap -1040 256 R0
SYMATTR InstName C2
SYMATTR Value 6.8n
SYMBOL res -1040 -112 R0
SYMATTR InstName R1
SYMATTR Value 39K
SYMBOL res -1440 240 R0
SYMATTR InstName R2
SYMATTR Value 22K
SYMBOL ZZZ\\potentiometer\\potentiometer\ 2 -1216 -16 R90
SYMATTR InstName U1
SYMATTR Value Rt=100K set=1
SYMBOL cap 256 -128 R0
SYMATTR InstName C3
SYMATTR Value 22n
SYMBOL cap 656 208 R0
SYMATTR InstName C4
SYMATTR Value 22n
SYMBOL res 656 -144 R0
SYMATTR InstName R4
SYMATTR Value 39K
SYMBOL res 256 192 R0
SYMATTR InstName R5
SYMATTR Value 8.2K
SYMBOL ZZZ\\potentiometer\\potentiometer\ 2 416 -48 M90
SYMATTR InstName U2
SYMATTR Value Rt=100K set=1
SYMBOL voltage -2192 48 R0
WINDOW 123 24 44 Left 2
WINDOW 39 0 0 Left 0
SYMATTR Value2 AC 1
SYMATTR InstName V1
SYMATTR Value SINE(0 .75 1000)
SYMBOL res 112 -160 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R10
SYMATTR Value 1.2K
SYMBOL res -1936 -416 R0
SYMATTR InstName R9
SYMATTR Value 3.3Meg
SYMBOL res -1936 -128 R0
SYMATTR InstName R11
SYMATTR Value 2.2Meg
SYMBOL res -1840 -128 R0
SYMATTR InstName R12
SYMATTR Value 22K
SYMBOL voltage -1360 -464 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V2
SYMATTR Value 9
SYMBOL cap -112 -112 R0
SYMATTR InstName C6
SYMATTR Value 2.2µ
SYMBOL res 640 -48 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R13
SYMATTR Value 47K
SYMBOL res -1296 -16 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R14
SYMATTR Value 39K
SYMBOL cap -1600 -528 R0
SYMATTR InstName C7
SYMATTR Value 22µ
SYMBOL ZZZ\\OpAmp\\TL072 -336 144 R0
SYMATTR InstName U3
SYMBOL cap -496 144 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C9
SYMATTR Value 22n
SYMBOL res -448 -224 R0
SYMATTR InstName R17
SYMATTR Value 10K
SYMBOL res -240 -64 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R18
SYMATTR Value 33k
SYMBOL cap 784 80 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C10
SYMATTR Value 22n
SYMBOL res 832 -288 R0
SYMATTR InstName R19
SYMATTR Value 10K
SYMBOL res 1040 -128 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R20
SYMATTR Value 33k
SYMBOL res -480 224 R0
SYMATTR InstName R24
SYMATTR Value 2.2Meg
SYMBOL res 800 80 R0
SYMATTR InstName R25
SYMATTR Value 2.2Meg
SYMBOL res -1152 -608 R0
SYMATTR InstName R15
SYMATTR Value 27K
SYMBOL res -1152 -464 R0
SYMATTR InstName R16
SYMATTR Value 27K
SYMBOL cap 1168 64 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C8
SYMATTR Value 4.7µ
SYMBOL cap -1248 -480 R0
SYMATTR InstName C11
SYMATTR Value 22µ
SYMBOL ZZZ\\potentiometer\\potentiometer\ 2 -800 96 R0
SYMATTR InstName U5
SYMATTR Value Rt=50K set=1
SYMBOL ZZZ\\OpAmp\\TL072 944 80 R0
SYMATTR InstName U4
SYMBOL res 1408 160 R0
SYMATTR InstName R7
SYMATTR Value 1Meg
SYMBOL res 1344 64 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R8
SYMATTR Value 470
SYMBOL res -1488 -128 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R21
SYMATTR Value 1.2K
SYMBOL cap -1504 16 R0
SYMATTR InstName C12
SYMATTR Value 15n
SYMBOL cap -1664 -128 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C13
SYMATTR Value 2.2µ
SYMBOL cap -1968 -176 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C5
SYMATTR Value 22n
SYMBOL njf -1872 -224 R0
SYMATTR InstName J1
SYMATTR Value J113-real
TEXT -1816 408 Left 2 !.Ac oct 100 40 10000
This file doesn't include the 100n power supply bypass cap. I've never used Kicad so I'm not sure if this is what you're looking for- let me know if you want something different.
|
|
|
|
Post by thetragichero on Jun 19, 2022 20:37:36 GMT -5
don't know how to easily import into kicad so i'm putting the schematic in manually. suppose it's good practice learning a new tool. will be exciting if i can start to make my own pcbs
|
|
|
|
Post by thetragichero on Jun 20, 2022 11:01:02 GMT -5
laying out something this complicated to pcb for my first attempt is a bit daunting, and now that i see physical component size i think it might be a bit too big for control cavity
but the general idea got me to thinking: buffer -> some sort of tilt/big muff tone control -> switch to put it into either the inverting or non-inverting op amp input to get either a mid cut or mid boost (although i suppose this would need three poles to switch the RC and CR filters to opposite ends)
still kicking stuff around obviously
|
|
|
|
Post by thetragichero on Jun 20, 2022 20:39:27 GMT -5
this is what i came up with. would've liked flat mids to be a few db lower but oh well    now to try to work up a board layout. any issues seen please let me know! |
|
|
|
Post by Yogi B on Jun 21, 2022 0:10:52 GMT -5
this is what i came up with. would've liked flat mids to be a few db lower but oh well Using a larger tone pot and/or lowering R8 would lower the volume at the centre of the sweep. If you want this to apply only when the flat mids selected, obviously this would need to be added to the toggle switch (this might not necessarily mean the requirement of an extra pole). In order to match the flat mids plot you posted, it seems that R5 should be 33k not 10k. Also I don't see the point of R3, plus at its present value of 10k it'll interfere with the tonestack — it should to be larger, 100k or more, to prevent that. If it is even needed at all.Edit: that last part is the consequence of my brain pre-breakfast, please ignore. (Though, to be clear, I still don't see the point of R3.) |
|
|
|
Post by thetragichero on Jun 21, 2022 9:06:15 GMT -5
you are correct and i have fixed my schematic
|
|
|
|
Post by Yogi B on Jun 21, 2022 16:31:34 GMT -5
Here's a further tweaked version, and although some values are changed it's the same number/type of components (however the toggle switch is upped to a DPDT): Which results in the following curves (red: flat-mids, yellow: scooped-mids): Changes: - As mentioned above Sw1 is now a DPDT (it is still depicted in the flat-mids position).
- C1 upped to 47n, this makes the flat response no lower than −1dB down to approx. 36Hz (with 22n it's 66Hz, which would be fine for standard tuning, but based on your mid-notch frequency I assume you're downtuning?).
- R3 (15k) & C4 (68n) have been repurposed to adjust the LPF in flat mode, somewhat upping the cutoff frequency & keeping the level of bass boost roughly consistent. In scooped-mids their effect should be minimal, though there could still be a better way to switch them out of the circuit.
- Additionally, since C4 was previously the coupling cap, I should mention that such a cap should no longer be required — the only voltage DC coupled to the tone stack is Vref.
- R6 is upped to 5k6 to compensate for the fact that it is switched in parallel with R5 (33k), rather than replacing it. (5k6 and 33k in parallel = 4k79.)
- R7 (33k) is repurposed as an extra loading resistor, switched in parallel with R8 (220k) when set to flat-mids.
- R9 is raised to 36k in order to lower the gain of the recovery stage and overall result in approximately unity gain at the flat setting.
- Oh, and finally, I've swapped the tone pot (RV1) to be the 'correct' way around — treble-boost/bass-cut when turing towards "10".
|
|
|
|
Post by thetragichero on Jun 21, 2022 22:16:37 GMT -5
figured out pcb routing in kicad and have installed the windows version of flatcam (linux version kept giving me errors related to some python-pyqt4 package that does not seem to exist. i have pyqt5 but even changing the python script to reference pyqt5 threw the same error. so windows32 version in wine it is) to turn my gerber and drill files into gcode. got my board and drill lined up and mirrored but i can't make the g code until i get the machine settings from my dad  i'll note the change to 47nf whenever i make and populate the board got the bridge pickup installed (straight to jack) and temporarily soldered just to test, guitarcenter order arrived with a bunch of packs of strings. had a $10 off coupon so i got a bunch of things to try on various guitars. this one got dr ddt 12-60 tuned to c# minor chord. guitar is setup and i'm digging how this plays for a $3-400 new guitar, sounds killer into the modified jet city picovalve i keep in the shop (amp is a little darker than i'd like but that means when i push it with a dirt pedal it sounds really really nice) will post a picture here from my phone. the neck right now is an empty pickup shell... whenever i figure out where my coil wire is packed that'll be a sustainer with i guess another pcb. empty hole is getting a kill switch, will determine where my two switches go when it's time to install em (space will be at a bit of a premium) edit: and here's the guitar  edit 2: looking at the board it appears i may have lost a couple traces when i had to move them all from the front side (which i won't have) to the back side of the board. at least i've got part of the process figured out |
|
|
|
Post by thetragichero on Jun 21, 2022 23:35:28 GMT -5
okay much happier with this and the less white the faster this will cut  |
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jun 21, 2022 23:42:47 GMT -5
Still a few bits that are a bit too close for my eye site
Bottom left feels a bit near the pad
Top middle there is a ring around a square pad
Bottom right a earth point going via two pads near to each other when you have a bigger opening just to the left of it
Also there is a angle track , I'd do it near the opening to avoid errors.
Top right some pads look like they are just touching the track
Right middle there is a square pad and then a round one that doesn't seem the round pad is connected to any thing
|
|
|
|
Post by sumgai on Jun 21, 2022 23:55:41 GMT -5
trag,
Put R3 back on U1a's inverting input. Reason being, it's intended to minimize quiescent current, the effect of which is that when a signal comes in on the non-inverting input, a small but noticeable offset, sounding like a blip, can occur.... and usually does. This is due to the current mirror in the two input circuits that are never quite exactly the same, there's always some microvolt or microamp difference in there. Even at unity gain for this stage, the blip might become noticeable later in the circuit.
In U1b, you have the same thing going on, but here we see the feedback resistor working in tandem with the "calming" resistor to form a voltage divider, which in turn creates a form of gain control. You currently have it set to a gain factor of 10. But at these values, a fair amount of current is flowing through the op amp, and not doing anything for the good of the signal. I suggest that you replace them with something like Yogi suggested, a 100K or better for each half of the TL072, and for U1b, try a 820K or 1MΩ resistor. Less current consumed, about the same gain factor.
HTH
sumgai
|
|
|
|
Post by thetragichero on Jun 22, 2022 0:35:34 GMT -5
Still a few bits that are a bit too close for my eye site Top middle there is a ring around a square pad Right middle there is a square pad and then a round one that doesn't seem the round pad is connected to any thing these are either me not knowing kicad well or whatever, the 9v input has the ring around the pad and the two unconnected pads are the signal input |
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jun 22, 2022 5:52:04 GMT -5
so is this going to be done by CNC or Etching , ever way you only use the TOP layer of a PCB can you not use the Easy EDA software and just use the Top layer Gerber File It lets you download the Gerber file so you can use it Easy EDADoing on CNC then you really want to fill in the gaps and just do like a Anti PCB i did work with a guy making PCB for me using a CNC (smart guy knew what i was doing making "Encoders") |
|
|
|
Post by Yogi B on Jun 22, 2022 6:07:28 GMT -5
Put R3 back on U1a's inverting input. Reason being, it's intended to minimize quiescent current Now that you mention it, I think I've previously read about this in some general op-amp article, somewhere. However I've never seen it discussed in a guitar-audio context, and as far as I'm aware I've only ever seen this done once (potentially by accident) and that's in the Klon. I'll also admit I was slightly getting ahead of myself, I was already thinking about tying R5/R6 to Vref: thus eliminating the DC path to ground (and arguably that, by itself, represents a bigger reduction in current draw) and simultaneously providing the same function as R3 (the combination of R4, RV1 & R5/R6 amount to a resistance of 133k7 or 166k between the inverting input and Vref). Finally, I don't know how trustworthy it is, but ltSPICE says that there'd be more current draw with R3 than without.
i'll note the change to 47nf whenever i make and populate the board With your version both the input cap (C1) and coupling cap (C4) would need to be larger — and since there are two, both need to be about double the 47n I used (i.e. 100n) in order to be equivalent. |
|
|
|
Post by thetragichero on Jun 22, 2022 10:16:01 GMT -5
thanks for all the suggestions folks. i've tidied up the layout and like this a lot better!  |
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jun 22, 2022 11:09:44 GMT -5
how you going to turn this on and off, because as it is and looks
even when the guitar is unplugged it will drain the battery
|
|
|
|
Post by thetragichero on Jun 22, 2022 11:21:38 GMT -5
active pickup so the battery ground isn't connected unless a cable is plugged into the trs jack. these are taking the place of the passive tone and volume that would normally be in the guitar (volume is still passive but you know what i mean)
|
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jun 22, 2022 12:41:52 GMT -5
Ah right yes, so the - on the battery hooks up with the Ground only when the Jack is in
I think i'll have a look at desiging this circuit at the weekend some thing to do.
and I'll try one for CNC (less cuts)
|
|
|
|
Post by sumgai on Jun 23, 2022 0:06:08 GMT -5
Finally, I don't know how trustworthy it is, but ltSPICE says that there'd be more current draw with R3 than without. That's exactly correct. But the thing is, the current you're speaking about here is that which is used to progress the signal long its path. In my original post, I was speaking of the quiescent current. Your idea of increased resistance is not only good, it's pretty much in line with the original chip designers' specs. It's, once again, a trade-off - more current draw versus a potential artifact disrupting the signal. I've not done much research of low-frequency (audio) op-amps in the past few years, but even a decade ago, there were much better quality op-amps available. I'm sure that the state of the art is by now quite capable of dropping a substitute part in place of that TL072, and making the (R3) resistor more-or-less unnecessary. But speaking of trade-offs..... at what cost? HTH sumgai |
|
|
|
Post by thetragichero on Jun 23, 2022 8:29:44 GMT -5
currently fighting with cnc that seems to be more interested with dive-bombing its bit into the board than cutting it. might be awhile to have enough patience to troubleshoot
(in the interim I've almost completely cleaned up a gibson br-9 lapsteel and amp so i should get around to making a video of it today)
|
|
|
|
Post by Yogi B on Jun 25, 2022 10:04:45 GMT -5
Finally, I don't know how trustworthy it is, but ltSPICE says that there'd be more current draw with R3 than without. That's exactly correct. But the thing is, the current you're speaking about here is that which is used to progress the signal long its path. In my original post, I was speaking of the quiescent current. Actually I meant both, quiescent or otherwise. As you previously mentioned, even with no input the op-amp output will be a few mV different to Vref, so connecting a resistor between the two dissimilar voltages will obviously cause a current to flow. LTspice reckons 330nA with the 10k resistor (i.e. a 3.3mV difference).
I've been reading around, looking for an explanation as to the cause of effects you describe, or at least acknowledgement that this specific issue exists — so far that search has come up lacking. What I have seen is the problem of needing to account for input bias current (I think this is actually what I was remembering having previously read about). The bias current causing additional voltage drop to the input signals thus increasing the offset of the output voltage. As the TL07x family has FET inputs, thus relatively low input bias current (65pA @ 25°c), this effect isn't that major (even if we had 1Meg source impedance, the voltage caused by input current would be 65pA × 1MΩ = 65μV). Also, notably, the 'fix' isn't: to provide 'extra' current via a resistor (with your implication that a lower resistor would always yield better performance, at the cost of greater current draw), but rather to match this voltage drop for both inputs; and with a non-inverting buffer the way to do that across the whole signal cycle isn't connecting a resistor from the inverting input to an AC ground (e.g. Vref), but rather connecting the resistor in the feedback loop. (Nor have I found any mention that this problem could lead to audible artefacts of the kind you describe.) An alternative issue I saw mentioned is that when driving heavy loads, the output of some op-amps may degrade to class-B (or worse) operation, causing cross-over distortion. This seems to better fit your description of a "blip" occurring upon the onset of an input signal. The suggested solution for this is to essentially force the op-amp into class-A operation via drawing significant output current even in the quiescent state. And the simplest way to achieve that obviously being a resistor — but a pull-up/pull-down resistor, connected from the output to either of the power rails, i.e. not connected to Vref. |
|
