When I called to order some inductors a few weeks back, I was saddened to learn that Bel had passed away a year or so ago. Irene said there wasn't anyone there who could answer technical questions. I guess that makes Scott the worlds leading authority on Opamp Labs gear.
I wanted to build up a couple of quasi-109 eqs from some 105P & 106P kits plus some extra 8709-3 inductors. Since I always want to learn something from every DIY project, I was looking closely at the schematic, which is in the old print catalogs, but not on their website. It can be found several places online, including here:
What I noticed is that when the low frequency control is switched to shelving (caps out of the circuit), it appears that the 50 & 100 Hz settings are using the exact same RL values. Same thing with the 300 & 400 Hz settings. Similarly, on the high end (with the inductors switched out), the 7 & 10 KHz settings are using the exact same RC values, as are the 12.5 & 15 KHz settings.
Is this a mistake on the schematic, or am I missing something?
>When I called to order some inductors a few weeks back, I was saddened to >learn that Bel had passed away a year or so ago. Irene said there wasn't >anyone there who could answer technical questions. I guess that makes >Scott the worlds leading authority on Opamp Labs gear.
Oh, that's not good.
That's a shame too. Bel was a great guy, and always willing to talk about stuff. He wasn't always so willing to write things down.
>I wanted to build up a couple of quasi-109 eqs from some 105P & 106P kits >plus some extra 8709-3 inductors. Since I always want to learn something >from every DIY project, I was looking closely at the schematic, which is >in the old print catalogs, but not on their website. It can be found >several places online, including here:
>What I noticed is that when the low frequency control is switched to >shelving (caps out of the circuit), it appears that the 50 & 100 Hz >settings are using the exact same RL values. Same thing with the 300 & >400 Hz settings. Similarly, on the high end (with the inductors switched >out), the 7 & 10 KHz settings are using the exact same RC values, as are >the 12.5 & 15 KHz settings.
>Is this a mistake on the schematic, or am I missing something?
Nope, that's right. The -6dB points on the filters are going to move somewhat. When you take one pole away, the other pole moves a little because the components aren't anything like perfect. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis."
On Thu, 05 Nov 2009 11:31:30 -0500, Scott Dorsey wrote: > Agent 86 <maxwellsm...@control.gov> wrote:
>>I wanted to build up a couple of quasi-109 eqs from some 105P & 106P >>kits plus some extra 8709-3 inductors. Since I always want to learn >>something from every DIY project, I was looking closely at the >>schematic, which is in the old print catalogs, but not on their website. >> It can be found several places online, including here:
>>What I noticed is that when the low frequency control is switched to >>shelving (caps out of the circuit), it appears that the 50 & 100 Hz >>settings are using the exact same RL values. Same thing with the 300 & >>400 Hz settings. Similarly, on the high end (with the inductors >>switched out), the 7 & 10 KHz settings are using the exact same RC >>values, as are the 12.5 & 15 KHz settings.
>>Is this a mistake on the schematic, or am I missing something?
> Nope, that's right. The -6dB points on the filters are going to move > somewhat. When you take one pole away, the other pole moves a little > because the components aren't anything like perfect. --scott
That makes sense, but I'm not sure we're talking about the same thing. I understand the frequency may change when switching from peaaking to shelving (taking one pole away).But I can't figure out how a .01uF cap with a 1.5k resistor makes a 7KHz filter in one place, but the same combination makes a 10kHz filter a quarter inch away. Unless the inductors are still affecting the circuit somehow. Granted, it's a shunt and not a "true bypass", but I'd expect an inductor (or capacitor) in parallel with a straight wire to behave pretty much like a straight wire. Is that not the case?
Can you recommend a resource that's easily available and doesn't presume too much prior knowledge to learn more about this stuff?
> On Thu, 05 Nov 2009 11:31:30 -0500, Scott Dorsey wrote: > > Agent 86 <maxwellsm...@control.gov> wrote:
> >>I wanted to build up a couple of quasi-109 eqs from some 105P & 106P > >>kits plus some extra 8709-3 inductors. Since I always want to learn > >>something from every DIY project, I was looking closely at the > >>schematic, which is in the old print catalogs, but not on their website. > >> It can be found several places online, including here:
> >>What I noticed is that when the low frequency control is switched to > >>shelving (caps out of the circuit), it appears that the 50 & 100 Hz > >>settings are using the exact same RL values. Same thing with the 300 & > >>400 Hz settings. Similarly, on the high end (with the inductors > >>switched out), the 7 & 10 KHz settings are using the exact same RC > >>values, as are the 12.5 & 15 KHz settings.
> >>Is this a mistake on the schematic, or am I missing something?
> > Nope, that's right. The -6dB points on the filters are going to move > > somewhat. When you take one pole away, the other pole moves a little > > because the components aren't anything like perfect. --scott
> That makes sense, but I'm not sure we're talking about the same thing. I > understand the frequency may change when switching from peaaking to > shelving (taking one pole away).But I can't figure out how a .01uF cap > with a 1.5k resistor makes a 7KHz filter in one place, but the same > combination makes a 10kHz filter a quarter inch away. Unless the > inductors are still affecting the circuit somehow. Granted, it's a shunt > and not a "true bypass", but I'd expect an inductor (or capacitor) in > parallel with a straight wire to behave pretty much like a straight wire. > Is that not the case?
> Can you recommend a resource that's easily available and doesn't presume > too much prior knowledge to learn more about this stuff?
> Thanks.
Agent,
your analysis is correct..
the same parts values will yield the same the same freq response...
so either the schematic is in error
or the device has the same response in the SHELVING mode in the 7 kHz and 10 kHz positions and in the other cases you noted.
The small inductance of the wire and switch will have no impact at audio freqs...
A circuit simulation program such as PSPICE or any others would allow you to explore this.. For this purpose you don't need to use a real opamp simulation, you can se an ideal op amp
>That makes sense, but I'm not sure we're talking about the same thing. I >understand the frequency may change when switching from peaaking to >shelving (taking one pole away).But I can't figure out how a .01uF cap >with a 1.5k resistor makes a 7KHz filter in one place, but the same >combination makes a 10kHz filter a quarter inch away. Unless the >inductors are still affecting the circuit somehow. Granted, it's a shunt >and not a "true bypass", but I'd expect an inductor (or capacitor) in >parallel with a straight wire to behave pretty much like a straight wire. >Is that not the case?
Think about it this way.
When we measure a peaking filter, we measure the center frequency.
When we measure a shelving filter, we measure the -6dB point.
Even if the poles _are_ in the same place, the point where we measure the filter from might not be the same.
But no, these inductors aren't all that wonderful. Capacitors are generally more like theoretical capacitors than inductors are like theoretical inductors.
>Can you recommend a resource that's easily available and doesn't presume >too much prior knowledge to learn more about this stuff?
Tremaine's Audio Cyclopedia? --scott
-- "C'est un Nagra. C'est suisse, et tres, tres precis."
>That makes sense, but I'm not sure we're talking about the same thing. I >understand the frequency may change when switching from peaaking to >shelving (taking one pole away).But I can't figure out how a .01uF cap >with a 1.5k resistor makes a 7KHz filter in one place, but the same >combination makes a 10kHz filter a quarter inch away. Unless the >inductors are still affecting the circuit somehow.
The inductor is turning it from a shelving filter into a peaking filter.
>Granted, it's a shunt >and not a "true bypass", but I'd expect an inductor (or capacitor) in >parallel with a straight wire to behave pretty much like a straight wire. >Is that not the case?
It is absolutely the case. Are we looking at the same schematic? --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis."
On Sun, 08 Nov 2009 13:50:41 -0500, Scott Dorsey wrote:
> The inductor is turning it from a shelving filter into a peaking filter.
Right, but my question is only about the shelving side of the filter. With the inductors bypassed, the 10k & the 7k filters both consist of just a .01uF cap & a 1.5k resistor. That's what I'm having trouble with.
> Are we looking at the same schematic?
I'm beginning to wonder about that myself. The link im my original post points to the same diagrahm that's on page 23 of both of the print catalogs I have here. These caataalogs are marked 3 JAN 97 at the top of each page.
Agent 86 wrote: > On Sun, 08 Nov 2009 13:50:41 -0500, Scott Dorsey wrote: >> The inductor is turning it from a shelving filter into a peaking filter.
> Right, but my question is only about the shelving side of the filter. > With the inductors bypassed, the 10k & the 7k filters both consist of > just a .01uF cap & a 1.5k resistor. That's what I'm having trouble with.
>> Are we looking at the same schematic?
> I'm beginning to wonder about that myself. The link im my original post > points to the same diagrahm that's on page 23 of both of the print > catalogs I have here. These caataalogs are marked 3 JAN 97 at the top of > each page.
I suspect the capacitor in the 7K shelving EQ should be 0.015uF. Maybe the tolerance was so wide in the original caps that they were hand selected from a bunch of 0.01uF caps to use in one position or the other. Or maybe it's a misprint (more likely).
|
