Thank you. I asked for formulas because I'd like to compute the range of possible resistor setting for achieving a certain freq-cut given the choice of the base caps.
The II order is a Sallen-Key toppology and the IV order is a pair of cascaded Sallen-Key with Q=0.5, am I correct?
In that case the formulas are pretty simple. I like the way is done here on wikipedia
https://en.wikipedia.org/wiki/Sallen–Key_topology
Thanks a lot!
The II order is a Sallen-Key toppology and the IV order is a pair of cascaded Sallen-Key with Q=0.5, am I correct?
In that case the formulas are pretty simple. I like the way is done here on wikipedia
https://en.wikipedia.org/wiki/Sallen–Key_topology
Thanks a lot!
In addition, to that app, the following one from Elliot Sound Products allows you to enter the desired crossover point and either a resistor or capacitor value and calculates the remaining value. I find it much easier than randomly entering values until you get something close to what you want.
https://sound-au.com/project09.htm#download_esp_lr
My comments were about lowering LF separation in the cutting step, to compensate for a recording that was going to generate too much vertical excursion. Using it in the playback step, to treat rumble or warp frequencies, is a different matter.
Using this technique in playback was well discussed in the threads on the Wyn Palmer designed phono stage, which incorporated the feature. You can find one of the threads here on DIYaudio:
Here
But more of the discussion happened over on Audiokarma:
audiokarma.
You might find the explanations there will answer your question. BTW I built a couple of his phono stages, and like them, but I left out that part of them because I'm not interested in narrowing the soundfield at LF in my system. I don't have a rumble problem, and if I have a record that's too warped to play I flatten it. What I do have is a lot of experience with narrowing the soundfield at LF when I have to, and I have never liked it. If a recording has spaciality down to the lowest frequencies, I find I like listening to it more untreated than treated. Other people's views may differ.
The discussion stems from a DIY biamp crossover builder's question about playing warped vinyl on an active system that has has a low enough crossover point (80 and 120Hz in our cases) that the lower channels can be summed to mono in order the make the vinyl more playable.
The question is how much, if any, stereo signal is lost when doing so, and in trying to get a handle on it, I've looked into this a bit over the last few days. I've learned or confirmed that;
- The inherent limitations in vinyl require the stereo to be blended out at LF.
- It seems generally thought best practice to blend to full mono by about 150Hz, and to gradually increase the amount of stereo above this.
- This blending is most often done further upstream in the recording and mixing process before the vinyl mastering.
- If there is still excess stereo in the final mix, this can be taken out at mastering.
- a few high end vinyl houses advertise of keeping as much LF stereo as possible where the music doesn't require too much volume (e.g. In an orchestral recording, a double bass can be hard panned to the right if the passage is quiet enough).
I've also done a few initial playback tests and measurements myself, and at my present level of ignorance, have not been able discern any meaningful stereo info below 120Hz. I haven't yet tested this with orchestral music.
I'm not able to reliably notice any difference between listening with the warp summed out or not (even when playing a non-warped record), although I myself would not want to leave it in when not needed, some people do.
With the LF summing to mono at 120Hz (24dB/oct) the sense of LF spaciality is preserved.
Overall therefore, the crossover builder can easily decide for themselves whether the LF sum to mono trick works for the intended purpose, or if you do indeed loose too much spaciality.
Probably these questions have been addressed before but I probably missed to spot it given the huge discussion.
Question 1
The circuit has a 0.1uF cap in the input buffer, but this is labeled as "10" on the board, which one is correct?
Question 2
Do I understand correctly that the circuit involves two polarized electrolytic capacitors in input and output? At least that is what I understand from reading this BOA. Now, I am in no way going to debate the design choices of a legend like Nelson, but I don't sleep at night with it. Is it possible to take the input section (50K pot in series with the cap) and the output section (starting before the output cap) off the PCB so that physically larger capacitors can be used? Is there any contraindication to leaving the PCB leads in those places?
Question 3
And we come to the input/output capacitors. Is it possible to use smaller film capacitors such as the usual 3-5uF of many tube line stages?
Question 4
Since I use a tube preamplifier that does not digest low impedances, I'd like to have >50K in input. Can I replace the two input pots with 250K pots or fixed resistors?
Question 5
For the possible input potentiometer, is it better to use a linear or logarithmic one? My intuition would suggest a linear potentiometer in this case.
Question 1
The circuit has a 0.1uF cap in the input buffer, but this is labeled as "10" on the board, which one is correct?
Question 2
Do I understand correctly that the circuit involves two polarized electrolytic capacitors in input and output? At least that is what I understand from reading this BOA. Now, I am in no way going to debate the design choices of a legend like Nelson, but I don't sleep at night with it. Is it possible to take the input section (50K pot in series with the cap) and the output section (starting before the output cap) off the PCB so that physically larger capacitors can be used? Is there any contraindication to leaving the PCB leads in those places?
Question 3
And we come to the input/output capacitors. Is it possible to use smaller film capacitors such as the usual 3-5uF of many tube line stages?
Question 4
Since I use a tube preamplifier that does not digest low impedances, I'd like to have >50K in input. Can I replace the two input pots with 250K pots or fixed resistors?
Question 5
For the possible input potentiometer, is it better to use a linear or logarithmic one? My intuition would suggest a linear potentiometer in this case.
Answer 1:
0.1uF = 100nF will give you a low at 1.6 Hz - low enough I will think. I must admit I used 10uF polypropylene and both in- and outputs. I did not notice until later.
Especially for the high pass section 0.1uF should suffice.
Answer 2:
As mentioned I used PP, however on my Korg preamp I used the electrolytics, and I have a sound(!) sleep.
I installed my 10uF PP caps with wiring to and from the PCB, and there might be some pictures earlier in this thread.
Answer 3:
For the input capacitors as mentioned even 100nF should be OK in my opinion. The output capacitors "work into" 47k however, and further on into the power amplfiers input impedance. 2.2uF should into 47k should give you a -3 dB at 1.5 Hz. If the power amplifier has 10k input impedance the 2.2uF will land the -3dB point at below 10 Hz.
Answer 4:
I have not tried this, but I guess it is a fair change it will be OK. You have to try. It might pick up more noise.
Answer 5:
No experience. But one thing I will say: I am using this crossover with Quads and SVS subwoofers, and I regret I used trim pots at the input because I am adjusting on the subwoofer. Please not in other cases it might be the higher frequencies, not the bass you want to attenuate in which case the crossover or the power amplifier for the higher frequencies needs a potentiometer.
That concludes my views and experience on this - good luck to you with the experiments.
0.1uF = 100nF will give you a low at 1.6 Hz - low enough I will think. I must admit I used 10uF polypropylene and both in- and outputs. I did not notice until later.
Especially for the high pass section 0.1uF should suffice.
Answer 2:
As mentioned I used PP, however on my Korg preamp I used the electrolytics, and I have a sound(!) sleep.
I installed my 10uF PP caps with wiring to and from the PCB, and there might be some pictures earlier in this thread.
Answer 3:
For the input capacitors as mentioned even 100nF should be OK in my opinion. The output capacitors "work into" 47k however, and further on into the power amplfiers input impedance. 2.2uF should into 47k should give you a -3 dB at 1.5 Hz. If the power amplifier has 10k input impedance the 2.2uF will land the -3dB point at below 10 Hz.
Answer 4:
I have not tried this, but I guess it is a fair change it will be OK. You have to try. It might pick up more noise.
Answer 5:
No experience. But one thing I will say: I am using this crossover with Quads and SVS subwoofers, and I regret I used trim pots at the input because I am adjusting on the subwoofer. Please not in other cases it might be the higher frequencies, not the bass you want to attenuate in which case the crossover or the power amplifier for the higher frequencies needs a potentiometer.
That concludes my views and experience on this - good luck to you with the experiments.
Hello piecor,
I will try to answer a few questions...:
Q1: I have used a 10µF bipolar cap (NICHICON UES BP - so called 'green sleeves') at the input and the output.
You could also use a PP-film cap there. The function of this cap is, that the virtual ground of the buffer
cells sits anywhere close to the middle between GND and close to ca. 21V (with 24 V rail - single ended).
Virtual ground is anywhere at +10 to +11 V , so you need a decoupling cap at the input and at the
output.
Q2: You can use physically larger caps at the input and at the output - if you want to lower the cutoff-
frequency (1 / 2xPi x R x C)
Q3: You could try a smaller filmcap at input / output. 3,3µF or 4.7 µF are worth a try. Don't ask me how it
sounds? This is your testfield.
Q4: You can try volume pots with higher values. Don't ask me how it behaves with a 250kOhm pot at the
input. Should be no problem?
Q5: For volume adjustments we normally use logarithmic pots. I used linear ones. And my ones are 10-turn
pots - possibility for very small adjustments. But that was my decision.
Hope this helps a little bit? Others are welcome to give a comment.
Cheers
Dirk
I will try to answer a few questions...:
Q1: I have used a 10µF bipolar cap (NICHICON UES BP - so called 'green sleeves') at the input and the output.
You could also use a PP-film cap there. The function of this cap is, that the virtual ground of the buffer
cells sits anywhere close to the middle between GND and close to ca. 21V (with 24 V rail - single ended).
Virtual ground is anywhere at +10 to +11 V , so you need a decoupling cap at the input and at the
output.
Q2: You can use physically larger caps at the input and at the output - if you want to lower the cutoff-
frequency (1 / 2xPi x R x C)
Q3: You could try a smaller filmcap at input / output. 3,3µF or 4.7 µF are worth a try. Don't ask me how it
sounds? This is your testfield.
Q4: You can try volume pots with higher values. Don't ask me how it behaves with a 250kOhm pot at the
input. Should be no problem?
Q5: For volume adjustments we normally use logarithmic pots. I used linear ones. And my ones are 10-turn
pots - possibility for very small adjustments. But that was my decision.
Hope this helps a little bit? Others are welcome to give a comment.
Cheers
Dirk
Thank you R-K Rønningstad!
Regarding question 3, I see your point. Actually one needs to consider the amp input Z in parallel with the 47K resistor, isn't it? In my case I have 47K input at the amp, therefore 47K parallel to 47K would be 23.5K. A 4uF cap would give the 0dB point at 16Hz (-3dB at 1.6Hz)!
I'm also thinking not installing the input pots at all. I don't want to attenuate the highpass and the lowpass will go through EQ correction anyway. Actually I'm also considering to replace the analog lowpass with a digital crossover that will allow me doing some EQ. I'll cross so low (80Hz region) that I don't care. I just wonder if a digital lowpass will be able to integrate perfectly with the 6-24 highpass... who knows!
I emailed the store, and this is the reply I received six days ago:
"Kits for the DIY Biamp 6-24 Crossover have been ordered in November 2024. The Helpdesk, however, does not have an estimated time for when new stocks will arrive.
Please sign up to get notified when the ACA kits become available by clicking the Notify me when available button here:"
"Kits for the DIY Biamp 6-24 Crossover have been ordered in November 2024. The Helpdesk, however, does not have an estimated time for when new stocks will arrive.
Please sign up to get notified when the ACA kits become available by clicking the Notify me when available button here:"
just click "Notify me" but if anyone wants to surrender this project then please give it to me 😁 pm me
New to this post. I am intersted to use this crossover for a 3 way active speakers. Actually I bought 2 kits yesterday.
I have browsed quickly through some of the posts but I could not find any info in regards to the phase response of the crossover. Has anyone measured or has any info on the phase response/transition at the crossover points and also over the whole spectrum? Thank you.
I have browsed quickly through some of the posts but I could not find any info in regards to the phase response of the crossover. Has anyone measured or has any info on the phase response/transition at the crossover points and also over the whole spectrum? Thank you.
If you use the simulator to pick your values it will show frequency response by amplitude and by phase.
The simulator has been moved from the location given in early pages of this thread. You can now find it here.
The simulator has been moved from the location given in early pages of this thread. You can now find it here.
David,
You have learned a lot that is overstated or just plain incorrect. Many things are written on the internet, some correct and some incorrect. Many things which are incorrect are repeated over and over. That doesn't make them correct.
Do you have the expertise to separate which is which?