Hi agdr,
That looks good. If or when you build one, could you please post your use and listening impressions? It would be nice to get some feedback, good or bad, so I can a better idea weather or not the eleven 1dB steps for a maximum of +/- 11dB is the best way to go, or would it be better to have something like .5dB steps with a +/- 5.5dB maximum, or maybe something in between? I've struggled a little with that part of the design, and it would be great to hear what others here think.
Mike
That looks good. If or when you build one, could you please post your use and listening impressions? It would be nice to get some feedback, good or bad, so I can a better idea weather or not the eleven 1dB steps for a maximum of +/- 11dB is the best way to go, or would it be better to have something like .5dB steps with a +/- 5.5dB maximum, or maybe something in between? I've struggled a little with that part of the design, and it would be great to hear what others here think.
Mike
i'd recommend using a pair of dual op amps instead of a single quad - you'd have a lot more options for "op amp rolling" if you so desire.
mlloyd1
mlloyd1 - Good thoughts but it may be a tight squeeze with 2 duals on each channel, since each one would need the pair of decoupling caps. Here are a couple of more options for audio quad chips with the same pinout, the OPA1664 (bipolar input) and OPA1644 (FET input):
http://www.ti.com/lit/ds/symlink/opa1664.pdf
http://www.ti.com/lit/ds/symlink/opa1644.pdf
A SOIC-8 to DIP-14 adaptor would be a slick way to use 2 duals for each quad (BrownDog #070401):
2 x Dual-to-Quad Op-Amp Adapter (p/n 070401)
but the way I have it laid out the input and one filter would be on one dual section, with the output and the other filter on the 2nd dual.
Hello Mike - I spotted your thread here when I was forum searching for a 2 band parametric tone control to potentially build, one where I could vary the corner frequencies as well as the amount of boost or cut. I've been put off by the use of pots in all the circuits I've found so far. I know how far off dual pots can be from section to section. It seems like a rotary switch would be a much better way to go. Have you ever messed around with a version of your circuit that would vary the bass and treble corner frequencies too (parametric)?
I see that Mouser has a 2 gang 23 postion rotary but it goes for $50 or so and they only have 1 in stock.
That PartsConnextion switch you found looks like a winner.I like being able to fit everything into off-the-shelf extruded boxes with card slots, Hammond or Box Enclosures, etc. The 80x100 would fit in several of the extruded aluminum boxes. I left 3mm free on each edge for card slots. They have 160x100 extruded cases that would also allow a 80mmx100mm power supply board to slide in behind the main board in the same slow. But... none of the extruded aluminum boxes look tall enough for the rotary switches. The Mouser switch is something like 2.25 inches in diameter. The alternalte would just be mounting the board in a larger box with standoffs, hence the 3mm corner holes.
I don't know if I'll fab/build the PC board or not, but I'll clean up that layout a bit and post the Eagle files so you or anyone else interested could modify them. I had to create custom parts in Eagle for the op amp and 3x9x11 resistor footprint. Those are relatively easy to add to the library in Eagle. The 80x100 size works with the free/DIY Eagle version but that is only two layers. I'll bet someone could condense it down to 2 layers using jumpers if necessary. I picked that size resistor footprint so that it works with either the tiny 3.9mm long 1/8W 1% metal film parts (the Xicon 270- series) or the 7mm long 0.1% tolerace metal film parts in that BOM. Those 0.1% are relatively low cost these days! I haven't calculated the power dissipation in the resistors, someone would need to do that. Given audio signal levels and the value of the resistors most should be well under 1/8W.
Last edited:
Mike,
I have been studying the latest version of your tone control and I believe it may provide an excellent basis for the preamp for my home audio system. I am interested in performing a few modifications, so I would like your input on what I propose to do as well as on a few technical questions. I thought it might help others who are also interested in your circuit if I post here instead of contacting you in a PM.
I would use the first inverting amplifier as a gain stage, as you have done. I would use an attenuator, either a potentiometer or a stepped switch with resistors, on the output of the second inverting amplifier. The output of the attenuator would feed switches which would provide the input for different power amplifiers controlling different speakers. I would use only one amplifier/speaker combination at a time. Each power amplifier would have FET input devices and 47,000 ohms impedance. Gain on these amplifiers would range from a low of 15dB on one to a high of 28dB on another. All speakers would have about 90dB sensitivity (2.83V at one meter). My preference is to listen in the close field at about one meter, and I would like to have 103dB peak output at that distance. My speakers have the capacity to produce that level if they have adequate input. I would use various sources such as phono preamp, CD player, cassette tape deck, radio tuner, and DAC output, all operating at the reference level of -10dBV. I hope to build the preamp into a single space rack unit enclosure. My questions:
1. Can you tell me the diameter of the rotary switch with resistors attached? I hope to fit it comfortably inside the 1U enclosure.
2. You use a 100kOhm resistor at the input to ground. Is this to give a high input impedance for any source?
3. The 10kOhm input resistor and the 20kOhm feedback resistor on the first inverting amplifier give a voltage gain of 2X (6dB). Is there any particular reason 20kOhm resistors were chosen for the feedback of the two inverting amplifiers?
4. Given the conditions I have noted above, I believe a preamp gain of 6dB should be adequate. Do you agree?
5. I would like to limit the bandwidth of the two inverting amplifiers to about 100,000 Hz by adding feedback capacitors in parallel with the feedback resistors. Is there any reason to not do so?
6. You use a 100 Ohm resistor at the output of the second inverting amplifier. I assume that this is to isolate that amplifier from any subsequent capacitance. Would it be wise to add the same value resistor to the output of the first inverting amplifier, or would that interfere with the function of the tone circuit?
7. Questions 5 and 6 are related to my desire to use the OPA627 op amp in all four postitions in your circuit, to ensure stability along with proper power supply bypassing at the op amp pins. Judging from the data sheet for the OPA627, I think it would work well in your circuit. Do you agree?
8. Can you (or other members) see any problems with what I am proposing to do?
Thanks again for posting you circuit, and thanks for any help you can give.
I have been studying the latest version of your tone control and I believe it may provide an excellent basis for the preamp for my home audio system. I am interested in performing a few modifications, so I would like your input on what I propose to do as well as on a few technical questions. I thought it might help others who are also interested in your circuit if I post here instead of contacting you in a PM.
I would use the first inverting amplifier as a gain stage, as you have done. I would use an attenuator, either a potentiometer or a stepped switch with resistors, on the output of the second inverting amplifier. The output of the attenuator would feed switches which would provide the input for different power amplifiers controlling different speakers. I would use only one amplifier/speaker combination at a time. Each power amplifier would have FET input devices and 47,000 ohms impedance. Gain on these amplifiers would range from a low of 15dB on one to a high of 28dB on another. All speakers would have about 90dB sensitivity (2.83V at one meter). My preference is to listen in the close field at about one meter, and I would like to have 103dB peak output at that distance. My speakers have the capacity to produce that level if they have adequate input. I would use various sources such as phono preamp, CD player, cassette tape deck, radio tuner, and DAC output, all operating at the reference level of -10dBV. I hope to build the preamp into a single space rack unit enclosure. My questions:
1. Can you tell me the diameter of the rotary switch with resistors attached? I hope to fit it comfortably inside the 1U enclosure.
2. You use a 100kOhm resistor at the input to ground. Is this to give a high input impedance for any source?
3. The 10kOhm input resistor and the 20kOhm feedback resistor on the first inverting amplifier give a voltage gain of 2X (6dB). Is there any particular reason 20kOhm resistors were chosen for the feedback of the two inverting amplifiers?
4. Given the conditions I have noted above, I believe a preamp gain of 6dB should be adequate. Do you agree?
5. I would like to limit the bandwidth of the two inverting amplifiers to about 100,000 Hz by adding feedback capacitors in parallel with the feedback resistors. Is there any reason to not do so?
6. You use a 100 Ohm resistor at the output of the second inverting amplifier. I assume that this is to isolate that amplifier from any subsequent capacitance. Would it be wise to add the same value resistor to the output of the first inverting amplifier, or would that interfere with the function of the tone circuit?
7. Questions 5 and 6 are related to my desire to use the OPA627 op amp in all four postitions in your circuit, to ensure stability along with proper power supply bypassing at the op amp pins. Judging from the data sheet for the OPA627, I think it would work well in your circuit. Do you agree?
8. Can you (or other members) see any problems with what I am proposing to do?
Thanks again for posting you circuit, and thanks for any help you can give.
Hi majerjack,
I'm going to have to go take a look before I can give you good answers, but right now I'm busy with other stuff. In the mean time here's my latest version that will actually be built.
I'll chime in later to answer your questions and reasons for the differences in the current version.
Mike
I'm going to have to go take a look before I can give you good answers, but right now I'm busy with other stuff. In the mean time here's my latest version that will actually be built.
I'll chime in later to answer your questions and reasons for the differences in the current version.
Mike
Hi all, I'm back.
First, the easy one.
Carl Huff, I'm using these - ConneX 2 Deck, 1 Pole/Deck x 24 Position, Silver-Plated, Shorting (MBB)I haven't been using them long enough to ascertain how they hold up long term, but they seem to be well constructed and I've not had any performance issues or problems so far. I did modify them to be 23 steps instead of 24 by placing a small peice of aluminum metal in the slot where the internal stop rides.
majerjack, here's the answers to your questions:
1) The switches are 1-1/2" in diameter including the terminals, and there's enough room to lay 1/4 watt resistors flat (so they aren't sticking out radially), so a 1U box might be a little tight, but probably workable.
2) The 100k resistor is just to provide a DC connection for the input side of the cap to prevent static charge, it's not at all necessary, it's there "just in case."
3) The 20k value of the feedback resistors is what I settled on as a compomise between lowest noise while keeping the internal circuit loading acceptable.
4) It really depends on things like the output level of your signal sources, the gain of the power amps, and the sensitivity of the speakers.
5) That really depends on which opamps are used and how well the circuit is designed and physically laid out, but I typically don't find them necessary, there's no need to address non-existant problems afterall. If the circuit is stable in all modes of operation (easily detectable with a decent o'scope), then they aren't needed.
6) Yeah, it's there to isolate against excessive capacitive loading. It's not needed after the first stage opamp, and would compromise the the performance of the tone filters, the resistors in the tone filters will provide that function.
7) The OPA627 would probably be OK, but they are somewhat more prone to instability in my experience, so I don't use them. They are also prohibitably expensive too, so unless you already have some on hand, I recommend you use a bipolar input device, and more specififcally the LME49720 (and its kin) is a real good chioce here.
8) I don't see any problem with your ideas for this, but as usual, YMMV.
Finaaly, for the new version I posted earlier, I came to the conclusion that 11dB of adjustment was way more than needed and re-designed it for a gain/cut of 6dB at 20Hz and 20kHz maximum. there's also an extra filter added to to the ouput stage for Baffle Step Compensation of my speaker system, it can be left out or frequency scaled as desired.
If there are other questions, Ill be happy to answer them as best I can.
Mike
First, the easy one.
Carl Huff, I'm using these - ConneX 2 Deck, 1 Pole/Deck x 24 Position, Silver-Plated, Shorting (MBB)I haven't been using them long enough to ascertain how they hold up long term, but they seem to be well constructed and I've not had any performance issues or problems so far. I did modify them to be 23 steps instead of 24 by placing a small peice of aluminum metal in the slot where the internal stop rides.
majerjack, here's the answers to your questions:
1) The switches are 1-1/2" in diameter including the terminals, and there's enough room to lay 1/4 watt resistors flat (so they aren't sticking out radially), so a 1U box might be a little tight, but probably workable.
2) The 100k resistor is just to provide a DC connection for the input side of the cap to prevent static charge, it's not at all necessary, it's there "just in case."
3) The 20k value of the feedback resistors is what I settled on as a compomise between lowest noise while keeping the internal circuit loading acceptable.
4) It really depends on things like the output level of your signal sources, the gain of the power amps, and the sensitivity of the speakers.
5) That really depends on which opamps are used and how well the circuit is designed and physically laid out, but I typically don't find them necessary, there's no need to address non-existant problems afterall. If the circuit is stable in all modes of operation (easily detectable with a decent o'scope), then they aren't needed.
6) Yeah, it's there to isolate against excessive capacitive loading. It's not needed after the first stage opamp, and would compromise the the performance of the tone filters, the resistors in the tone filters will provide that function.
7) The OPA627 would probably be OK, but they are somewhat more prone to instability in my experience, so I don't use them. They are also prohibitably expensive too, so unless you already have some on hand, I recommend you use a bipolar input device, and more specififcally the LME49720 (and its kin) is a real good chioce here.
8) I don't see any problem with your ideas for this, but as usual, YMMV.
Finaaly, for the new version I posted earlier, I came to the conclusion that 11dB of adjustment was way more than needed and re-designed it for a gain/cut of 6dB at 20Hz and 20kHz maximum. there's also an extra filter added to to the ouput stage for Baffle Step Compensation of my speaker system, it can be left out or frequency scaled as desired.
If there are other questions, Ill be happy to answer them as best I can.
Mike
Last edited:
For Michael Bean's circuit, scaling the capacitors up and down as practical will move the turnover frequency up and down accordingly, in a linear fashion. Bigger capacitors lower the turnover; smaller capacitors raise it.
A lot of equipment uses 1 kHz turnover frequencies for both bass and treble. Try spreading it out an octave or two. Lowering the bass turnover frequency below 600 Hz will eliminate some mid bass boost. Maybe try 300 Hz too. And raising treble turnover frequency above 2 kHz will eliminate some midrange boost. These are all things to explore with a breadboard and a few capacitors.
Walt Jung's classic shelving equalizer circuit is adaptable to this concept of switched resistors. If you can find it on the web (I couldn't but I have it in a couple of books "IC Op Amp Cookbook" and "Op Amps for Audio Applications") it contains cookbook equations to meet any practical design criteria.
Last edited:
Fast Eddie D,
If you take a look at the response curves you'll see that the HP and LP frequencies are already spread out as a consequence of using the accelerated slope filters, and that the mid-band frequencies aren't affected much at all. Having said that, if anyone wishes to modify the circuit to suit their needs, your advice is sound.
Mike
If you take a look at the response curves you'll see that the HP and LP frequencies are already spread out as a consequence of using the accelerated slope filters, and that the mid-band frequencies aren't affected much at all. Having said that, if anyone wishes to modify the circuit to suit their needs, your advice is sound.
Mike
Yes Mark, I didn't scrutinize the response curves. Your turnover frequencies - 300 Hz and 2 kHz - are the frequencies that I have determined empirically to be the most useful. Your circuit is much more useful than most consumer equipment - past and present - offers. It's very nice.
The trend is now away from tone controls and towards selectable contours - "Jazz, pop" etc plus a bass boost that is never useful. My buddy's mom bought a fancy Harman Kardon AV receiver that has these features, and they're useless.
My pleasure.
You might try a continuously variable loudness control. Nakamichi used it in their TA series receivers; you can look it up. It requires a tapped volume control, which as I pointed out is still available at Rat Shack.
The not so obvious beauty of a continuously variable loudness control is that when it's turned all the way up (no contour) it's completely out of the circuit.
Most (virtually all) loudness circuits of yesteryear introduced way too much compensation. The variable loudness control is so much better.
You will still need a conventional volume control.
You might try a continuously variable loudness control. Nakamichi used it in their TA series receivers; you can look it up. It requires a tapped volume control, which as I pointed out is still available at Rat Shack.
The not so obvious beauty of a continuously variable loudness control is that when it's turned all the way up (no contour) it's completely out of the circuit.
Most (virtually all) loudness circuits of yesteryear introduced way too much compensation. The variable loudness control is so much better.
You will still need a conventional volume control.
Generally, I think that this project is absolutely fantastic.
With a major goal of loudspeakers and amplifiers to have a flat response, despite the fact that human ears and ordinary rooms NEVER have a flat response, a high resolution device to make up the difference is absolutely vital, in my opinion.
With a major goal of loudspeakers and amplifiers to have a flat response, despite the fact that human ears and ordinary rooms NEVER have a flat response, a high resolution device to make up the difference is absolutely vital, in my opinion.