when i use the R-side for all resistors in the low pass, this should make the xo-frequency variable between about 54Hz and 220Hz.
for the high pass i need R and 2R.
what i calculated is an approximation (Q ~ 2R).
i made a quick plot to demonstrate what i did:
for the high pass i need R and 2R.
what i calculated is an approximation (Q ~ 2R).
i made a quick plot to demonstrate what i did:
You need to understand how it works before deciding on swapping components.
R1 determines the input impedance. So 22K here. The gain of the stage is (minus) -R2/R1. So it has a gain of -1 (it inverts). The input cap is essential with it being a single rail design. The value of the cap works in conjunction with R1 to determine lower frequency limit. So that is 0.7Hz for these two. If you change the 10uf input cap to 1uf then the -3db point rise to 7Hz. The value of the output cap works in conjunction with the load impedance the circuit is connected to.
If you reverted back to your original scheme (non inverting for input and adder) then the addition of the two signals would not be as accurate because one would modulate the other. The inverting configuration avoids that completely. It really is better. For the sake of four 10uf electroylitics and a few resistors I would stick with the inverting setup to be honest.
R1 determines the input impedance. So 22K here. The gain of the stage is (minus) -R2/R1. So it has a gain of -1 (it inverts). The input cap is essential with it being a single rail design. The value of the cap works in conjunction with R1 to determine lower frequency limit. So that is 0.7Hz for these two. If you change the 10uf input cap to 1uf then the -3db point rise to 7Hz. The value of the output cap works in conjunction with the load impedance the circuit is connected to.
If you reverted back to your original scheme (non inverting for input and adder) then the addition of the two signals would not be as accurate because one would modulate the other. The inverting configuration avoids that completely. It really is better. For the sake of four 10uf electroylitics and a few resistors I would stick with the inverting setup to be honest.
ah ok.
is there anything wrong with the variable freq?
or should i leave my cheapo pots in the stock-box, to avoid loss of sound quality?
is there anything wrong with the variable freq?
or should i leave my cheapo pots in the stock-box, to avoid loss of sound quality?
Looks like you have given the variable option some thought. Its difficult for me to say really... its making sense to you and you can see what you want.
If you want I can put the values into the LTspice simulation and see how the overall response looks but you would have to tell me the new resistor values (equivalent to a single resistor) to substitute... you have W8 though... do you want the file ?
If you want I can put the values into the LTspice simulation and see how the overall response looks but you would have to tell me the new resistor values (equivalent to a single resistor) to substitute... you have W8 though... do you want the file ?
Why not tag pots in and try it for real and see what settings actually sound the best in the room and for your setup. You can always substitute fixed values later.
two last questions for now:
you changed the values of the adder.
i thought it would be nice to have a bit of gain in that stage.
why is it wrong?
you have 100k resistors on the outputs.
do i need them or are they for simulation purposes?
you changed the values of the adder.
i thought it would be nice to have a bit of gain in that stage.
why is it wrong?
you have 100k resistors on the outputs.
do i need them or are they for simulation purposes?
Here is the file. Do you know how to run it LTspice and how to alter things ?
The 100k on the output is good practice. It defines ground for the cap and means that the cap does not develop a voltage when left floating... that would cause thumps and bangs when you plugged something into it. The input caps should really have the same too, a high value resistor across directly across the input to relate it to ground at all times.
You can change the gain of the "adder" or mixer" if you wish. It follows the same rule as the input stage for calculation.
The 100k on the output is good practice. It defines ground for the cap and means that the cap does not develop a voltage when left floating... that would cause thumps and bangs when you plugged something into it. The input caps should really have the same too, a high value resistor across directly across the input to relate it to ground at all times.
You can change the gain of the "adder" or mixer" if you wish. It follows the same rule as the input stage for calculation.
If you were retaining your original configuration then you would need an extra resistor to add gain to the "adder" stage. The value of R would set the gain. The cap would be around 22 to 47uf.
I still advise the inverting set up though 🙂
Attachments
the whole signal-ground thing is giving me a bad headache.
all stages except the low-pass have the zener-voltage as reference.
why can i use the 0V-potential of the batteries as signal-ground?
Would it even make a difference to completely not connect signal-ground to the preamp and instead bridge it from say a mp3-player into the power amplifier?
And mooly saying, it would also work to use the zener-voltage as signal-ground is too much for me.
I don´t even get why the lowpass needs 0v as ground.
Tommorow i make it to the store to buy the rest of the parts.
i have seen mkt caps with 10uF.
is there a difference in using elko or film caps for the in- and output?
is a 100k resistor to gnd not needed at the input?
all stages except the low-pass have the zener-voltage as reference.
why can i use the 0V-potential of the batteries as signal-ground?
Would it even make a difference to completely not connect signal-ground to the preamp and instead bridge it from say a mp3-player into the power amplifier?
And mooly saying, it would also work to use the zener-voltage as signal-ground is too much for me.
I don´t even get why the lowpass needs 0v as ground.
Tommorow i make it to the store to buy the rest of the parts.
i have seen mkt caps with 10uF.
is there a difference in using elko or film caps for the in- and output?
is a 100k resistor to gnd not needed at the input?
The Signal Ground is actually the Return wire of the Signal CIRCUIT.
The Signal Ground must be run as the partner to the Flow wire in the signal pair.
The Flow and Return PAIR allow the CIRCUIT to work. Omit one of the pair and you break the CIRCUIT.
The Signal Ground must be run as the partner to the Flow wire in the signal pair.
The Flow and Return PAIR allow the CIRCUIT to work. Omit one of the pair and you break the CIRCUIT.
The circuit in post #38 is correct and essentially complete. All that is missing is the 100uf cap across the zener and (recommended) the 100k across each input to ground. The 100k connects to the left hand side of the cap in the diagram.
In the diagram in post #38 the Vref is used just as a bias voltage for the opamps. It is not in the signal path. It is purely a reference voltage. This is a true and pure single rail design.
Personally I would be more than happy with good quality electrolytics for the input and output caps. Electrolytics have a bit of an undeserved reputation imo 🙂 Film caps in that size will be large and expensive.
everything is on the breadbord.
everything was checked three times.
a few seconds ago i started the first test.
...smoke...stink...zisch...knister...
now i know what a tl072 looks like from the inside.
immediately saw the mistake:
negative supply was connected to vref.
i only bought one tl072...so back to original design with only ne5532.
next test will start in few moments.
everything was checked three times.
a few seconds ago i started the first test.
...smoke...stink...zisch...knister...
now i know what a tl072 looks like from the inside.
immediately saw the mistake:
negative supply was connected to vref.
i only bought one tl072...so back to original design with only ne5532.
next test will start in few moments.
We've all done it 😀
If in doubt power up with no IC's fitted and then check supplies are correct on the appropriate pins.
If in doubt power up with no IC's fitted and then check supplies are correct on the appropriate pins.
i grilled 6 opamps over the weekend.
i have only one ne5532 left, five tl071, and five tl062.
already ordered a bunch of ne5532 and tl072 from china.
i must have done something totally wrong on the board.
i´ll try to build up stage after stage now with the tl071 left.
i´ll run a seperate test with each stage.
i have only one ne5532 left, five tl071, and five tl062.
already ordered a bunch of ne5532 and tl072 from china.
i must have done something totally wrong on the board.
i´ll try to build up stage after stage now with the tl071 left.
i´ll run a seperate test with each stage.
Yes, something is very wrong for that to happen. With correct supplies opamps are virtually bullet proof. You can pretty much short any and every pin to either rail or to anything else with no problem. China ? are you sure the IC's are what they say they are ?
i don´t know.
it is crazy.
i blame my cheap breadboard atm.
i have build up the input, adder, high with tl071 now.
works fine.
the sound is surprisingly good.
now i´ll try the low pass + subsonic with the remaining 5532.
edit.
no, will not do that.
i detected a bit of melted plastic above two pins of the 5532.
maybe i should try it with a tl062?
it is crazy.
i blame my cheap breadboard atm.
i have build up the input, adder, high with tl071 now.
works fine.
the sound is surprisingly good.
now i´ll try the low pass + subsonic with the remaining 5532.
edit.
no, will not do that.
i detected a bit of melted plastic above two pins of the 5532.
maybe i should try it with a tl062?
Last edited:
- Status
- Not open for further replies.