Author |
Message |
germaniac
Joined: Aug 04, 2006 Posts: 200 Location: California
Audio files: 7
|
Posted: Wed Jun 25, 2008 9:19 pm Post subject:
|
|
|
Hey JH,
Awaiting my PCBs, but looking at the schemo. Sorry if this is a dumb question, but what is the number next to each trimmer? (For instance, for filter #23 [top left corner], R TUNE resistance is 13K and the number beside the trimmer is .4772.) Also, what is the procedure for calibrating each filter?
Thanks for your help!
Regards,
Joe |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Wed Jun 25, 2008 11:57 pm Post subject:
|
|
|
germaniac wrote: | Hey JH,
Awaiting my PCBs, but looking at the schemo. Sorry if this is a dumb question, but what is the number next to each trimmer? (For instance, for filter #23 [top left corner], R TUNE resistance is 13K and the number beside the trimmer is .4772.) Also, what is the procedure for calibrating each filter?
Thanks for your help!
Regards,
Joe |
R_TUNE is the fixed resistor in series with the trimmer.
The trimmers are all 10k.
R_TUNE is chose such that, with all C's having their nominal values, each trimmer is almost in center position, allowing the best use of the trimmers to compenste for C tolerances in both directions. You will be able to trim out +/-10 percent C tolerances. (Worst case: both C's +10% or both C's -10%) By optimizing the circuit this way, you have an increadible fine resolution to adjust your resonance peaks. (About half a turn of the multiturn trimmers to just "ride the peak".)
The numbers like ".4772" are the actual trimpot setting for precise center frequency at nominal C values. Close to .5 (mid position), but of course not exactly 0.5 . Each trimpot is set to that "ideal" value in the schematics in order to get the required frequency response in PSpice.
These ".4772" numbers have no relevance to what setting you will end up with your board, as this depends on the actual C values (tolerances) you have.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
germaniac
Joined: Aug 04, 2006 Posts: 200 Location: California
Audio files: 7
|
Posted: Thu Jun 26, 2008 9:58 am Post subject:
|
|
|
jhaible wrote: | The numbers like ".4772" are the actual trimpot setting for precise center frequency at nominal C values. Close to .5 (mid position), but of course not exactly 0.5. |
Ah, so the ".4772" numbers indicate the rotation of the trimmer referenced to center. Got it, thanks!
As for calibrating then, did you tune each filter at individual test points, or from the master output, or . . . ? And do you use a test tone of some sort and/or frequency meter? In other words, what's your recommended procedure for tuning the filters?
Regards,
Joe |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Thu Jun 26, 2008 10:35 am Post subject:
|
|
|
germaniac wrote: | jhaible wrote: | The numbers like ".4772" are the actual trimpot setting for precise center frequency at nominal C values. Close to .5 (mid position), but of course not exactly 0.5. |
Ah, so the ".4772" numbers indicate the rotation of the trimmer referenced to center. Got it, thanks!
As for calibrating then, did you tune each filter at individual test points, or from the master output, or . . . ? And do you use a test tone of some sort and/or frequency meter? In other words, what's your recommended procedure for tuning the filters?
Regards,
Joe |
You connect a sine generator (*) to the input, set Q_POT to 12 o' clock position, feed the 40 nominal BPF frequencies one by one, and adjust for maximum level on the BPF that is in resonance.
(*) Sine genearator: I've used a synth + Frequency Counter function of my multimeter; could also be a DDS signal generator, or some software and PC soundcard.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
germaniac
Joined: Aug 04, 2006 Posts: 200 Location: California
Audio files: 7
|
Posted: Thu Jun 26, 2008 10:49 am Post subject:
|
|
|
jhaible wrote: |
You connect a sine generator (*) to the input, set Q_POT to 12 o' clock position, feed the 40 nominal BPF frequencies one by one, and adjust for maximum level on the BPF that is in resonance.
(*) Sine genearator: I've used a synth + Frequency Counter function of my multimeter; could also be a DDS signal generator, or some software and PC soundcard. |
Piece of cake! Thanks!
|
|
Back to top
|
|
|
elhardt
Joined: May 14, 2005 Posts: 73 Location: USA
|
Posted: Mon Jun 30, 2008 1:07 pm Post subject:
|
|
|
Inventor wrote: |
You told me to wait, so I did. The frequency plot helps a lot, but I didn't find the exact information I need which is: What is the center frequency and Q of each filter? With that Info i can quickly code something up! Thanks. |
If you want to get a slightly better idea of the filter freqs, I overlaid JH's plot onto an audio log graph at the link below. Since the filters appear evenly spaced logarithmically, if you can determine the spacing between two filters you can calculate the rest.
http://i127.photobucket.com/albums/p122/Elhardt/Misc/JHFilterPlot.jpg
I've written software that can filter wav files both in the frequency or time domains. It can do any shaped filter. If you get somebody who has built the JH string filter to run a single sample pulse through it and send me the resulting impulse file, I can provide quite accurate center frequencies and probably even an application that will allow you to filter wav files (maybe later real-time audio input) with JH's string filter.
-Elhardt |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Mon Jun 30, 2008 1:18 pm Post subject:
|
|
|
Hi Ken,
nice to see you here - welcome!
BTW, I have all the individual frequencies explicitely printed out in the documents I've published on my web site ...
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
elhardt
Joined: May 14, 2005 Posts: 73 Location: USA
|
Posted: Mon Jun 30, 2008 1:53 pm Post subject:
|
|
|
jhaible wrote: | Hi Ken,
nice to see you here - welcome!
BTW, I have all the individual frequencies explicitely printed out in the documents I've published on my web site ...
JH. |
Hello JH. Until a few weeks ago I didn't even realize there was this forum just for your designs.
As for the frequencies, I guess Inverter didn't know about those, so I assumed they weren't available online either. Thanks for the correction. |
|
Back to top
|
|
|
guitarfool
Joined: Feb 26, 2007 Posts: 160 Location: Maryland
Audio files: 8
|
Posted: Mon Jun 30, 2008 1:58 pm Post subject:
|
|
|
Well, I finished my JH String Filter yesterday. What seemed like the most tedious part was inserting all of those TL072's.
I was so anxious to get it up and running that I stopped work on th Frequency Shifter when I got these boards (it only needed mounting on a panel).
It actually sounded pretty good even before making any adjustments! Playing an electric guitar through it, I did notice a wolf note around D just above middle C, so it needs adjusting after all.
However, the wiring on the schematic for the stereo spread didn't look right to me. It shows the "r to l" on PCB1 and the "l to r" on PCB2 connectors having their pins 2 & 3 tied together. I swapped the connections to pins 2 & 3 on "r to l" (PCB1) and it seems to do what I think it's supposed to.
This is going to be the heaviest single-width Moog format module I have, weight wise |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Mon Jun 30, 2008 3:17 pm Post subject:
|
|
|
guitarfool wrote: |
However, the wiring on the schematic for the stereo spread didn't look right to me. It shows the "r to l" on PCB1 and the "l to r" on PCB2 connectors having their pins 2 & 3 tied together. I swapped the connections to pins 2 & 3 on "r to l" (PCB1) and it seems to do what I think it's supposed to.
|
These connectors are used for cross-feeding the two halves of the filterbank.
A look at the schematics may shed some light on how it works: the summing resistor that doe sthe cross-feeding is split into two parts, haolf of it sitting on the sending end , and half of it sitting on the receiving end. Then, a shunt (potentiometer + capacitor to GND) at the connection point decreases the amount of cross feeding.
"l_to_r" or "r_to_l" are short descriptions of the function of those connectors. As both partial filters use the same PCB, I had to decide on one labelling, which may be a bit confusing.
But in the end it's as easy as this: You connect pin 3 of PCB 1 to pin 2 of PCB 2 ("path 1->2"), and pin 3 of PCB 2 to pin 2 of PCB 1 ("path 2->1), and each of these connections gest its own variable shunt to GND.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
guitarfool
Joined: Feb 26, 2007 Posts: 160 Location: Maryland
Audio files: 8
|
Posted: Mon Jun 30, 2008 4:30 pm Post subject:
|
|
|
jhaible wrote: |
These connectors are used for cross-feeding the two halves of the filterbank.
A look at the schematics may shed some light on how it works: the summing resistor that doe sthe cross-feeding is split into two parts, haolf of it sitting on the sending end , and half of it sitting on the receiving end. Then, a shunt (potentiometer + capacitor to GND) at the connection point decreases the amount of cross feeding.
"l_to_r" or "r_to_l" are short descriptions of the function of those connectors. As both partial filters use the same PCB, I had to decide on one labelling, which may be a bit confusing. |
That's what I thought.
jhaible wrote: |
But in the end it's as easy as this: You connect pin 3 of PCB 1 to pin 2 of PCB 2 ("path 1->2"), and pin 3 of PCB 2 to pin 2 of PCB 1 ("path 2->1), and each of these connections gest its own variable shunt to GND.
JH. |
That's what I did. |
|
Back to top
|
|
|
alain6870
Joined: Apr 04, 2008 Posts: 17 Location: France
|
Posted: Sun Jul 06, 2008 12:44 pm Post subject:
parametric EQ for the string filter |
|
|
Hello Jürgen,
I've got my PCBs last week and I wanted to complete the filter with a parametric EQ as you did in your first version. I wonder if I get the same audio result as I put the EQ ( it would then be mono) before the filter or as it is placed after the filter (should then be stereo).
I presume that when the EQ is after the filter it is much easier to work with but it is more expansive to build.
Could you, please, give me a clue?
Thank you very much,
Alain |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Sun Jul 06, 2008 12:59 pm Post subject:
Re: parametric EQ for the string filter |
|
|
alain6870 wrote: | Hello Jürgen,
I've got my PCBs last week and I wanted to complete the filter with a parametric EQ as you did in your first version. I wonder if I get the same audio result as I put the EQ ( it would then be mono) before the filter or as it is placed after the filter (should then be stereo).
I presume that when the EQ is after the filter it is much easier to work with but it is more expansive to build.
Could you, please, give me a clue?
Thank you very much,
Alain |
In theory, it makes no difference whether the EQ is before or after the filter bank. But in practice - exactly as you say - you'd need two EQs if placed after the filter bank, so the only place the EQ should go is *before* the filter bank.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Mon Jul 14, 2008 10:49 pm Post subject:
|
|
|
String Filter PCBs are sold out.
(If you have pre-ordered this and not yet paid , your boards are still here and waiting for you.)
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
Stavros
Joined: Oct 30, 2007 Posts: 36 Location: Athens-Greece
|
Posted: Sat Jul 19, 2008 4:48 am Post subject:
|
|
|
Hello Jurgen,
I will appreciate few clarifications please(please excuse the stupid ones).
A. When utilizing a regulated +-15V regulated power supply
1. U1, U2, R1 through R6, D5, D6, C3 and C4 should not be used, correct?
2. +- 15V should be individually connected to PCB 1 and 2?
3. If yes, should capacitors C35 and C36 be soldered in?
4. If +-15V can be connected to PCB 1 only, then positions +0- of PCB 1 and 2 (close to Link and Link*) should be connected? Or should they be joined in any case?
B. Is “Link” simply connected to Link* on the second PCB?
C. P1 through P4 for the toggle switches are ON “when connected to 470R” or ON “OFF when connected to GND” correct?
D. Connectors “L to R” and “R to L” should be connected to a dual/parallel (Tandem) 100 k log potentiometer as shown on the schematic. Correct?
E. What is a recommended value for the “Input” potentiometer?
F. Should there also be a potentiometer for the Q-CV? If yes what is a recommended value?
G. Should one wish to connect the provided (40) single outs to potentiometers what would be the connection schema of one?
H. How are “Feed” and “Feed*” used?
I. How/Where would the forthcoming PCB for the Violin/Viola/Cello Preset EQ be connected to the String Filter PCBs?
J. Is it a viable EQ solution to use the “Performer Equalizer” of the “Triple Chorus Solina” as the first input of the “signal to be processed” and the output from the “Triple Chorus” as input to the “Strings Filter”?
K. Should a Front panel Design cover the forthcoming Preset filters and /or the Equalizer section?
Many thanks for your understanding and support
Best Regards
Stavros |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Sat Jul 19, 2008 7:19 am Post subject:
|
|
|
Stavros wrote: |
A. When utilizing a regulated +-15V regulated power supply
1. U1, U2, R1 through R6, D5, D6, C3 and C4 should not be used, correct?
|
Almost correct. leave C3 and C4 in place, but omit C1 and C2.
You don't need D1 thru D4 either.
I'm summarizing it:
For +/-15 V DC supply, omit
U1, U2,
R1 through R6,
D1 thru D6,
C1 and C2.
+- 15V should be connected to PCB 1 and 2
(You can use the two MOTM connectors, or you can use one MOTM connector and use the two small "+ 0 -" connectors to connect PCB 1 nad PCB 2, power-wise.)
Quote: |
3. If yes, should capacitors C35 and C36 be soldered in?
|
Yes.
Quote: |
B. Is “Link” simply connected to Link* on the second PCB?
|
Yes.
Quote: |
C. P1 through P4 for the toggle switches are ON “when connected to 470R” or ON “OFF when connected to GND” correct?
|
You can just use a simple on/off switch that closes the connection between the middle and right (470 Ohm) pin. The GND (left pin) is just there for a shield, if you have long cables.
But it doesn't hurt when you use an on/on switch that connects the middle pin to the left pin (GND) when "off".
Quote: |
D. Connectors “L to R” and “R to L” should be connected to a dual/parallel (Tandem) 100 k log potentiometer as shown on the schematic. Correct?
|
100k log tandem pot is right.
The schematics has the connection wrong (sorry), but the wiring diagram is right: http://www.jhaible.heim.at/string_filter/jh_string_filter_wiring.pdf
Quote: |
E. What is a recommended value for the “Input” potentiometer?
|
50k lin, 10k log, something like that. Not critical.
Quote: |
F. Should there also be a potentiometer for the Q-CV? If yes what is a recommended value?
|
It's an option. Value not critical. 50k lin would be ok.
More later. Gotta run.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Sat Jul 19, 2008 3:23 pm Post subject:
|
|
|
Stavros wrote: |
G. Should one wish to connect the provided (40) single outs to potentiometers
|
Well, it's a lot of potentiometers, and a lot of work to hook them all up, but yes, that's what the individual outs are for, if you want to go that route.
Quote: |
what would be the connection schema of one?
|
Each output goes to a 1k resistor (soldered in vertically), then to a 50k log potentiometer. If you want to get rid of any DC offset, put a 1uF capacitor between the resistor and the pot.
Then sum the potentiometer wipers with 51k resistors. Either to one summing point, or to two, for stereo. You can do passive summing and go to a sensitive input on the mixing desk, or hook up some opamps.
Quote: |
H. How are “Feed” and “Feed*” used?
|
One PCB had Feed, the other has Feed* (different location on PCB!).
Connect them, as you do with Link and Link*.
Quote: |
I. How/Where would the forthcoming PCB for the Violin/Viola/Cello Preset EQ be connected to the String Filter PCBs?
|
Input -> EQ -> Filter Bank -> Outputs.
Quote: |
J. Is it a viable EQ solution to use the “Performer Equalizer” of the “Triple Chorus Solina” as the first input of the “signal to be processed” and the output from the “Triple Chorus” as input to the “Strings Filter”?
|
It's nice, but it won't sound like a real string instrument.
BTW, If you have 40 sliders, you don't need no EQ at all: Just "draw" the EQ courve with the sliders.
JH.
Now playing: Paul Horn, Inside _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Sun Jul 27, 2008 6:40 am Post subject:
|
|
|
jhaible wrote: | String Filter PCBs are sold out.
(If you have pre-ordered this and not yet paid , your boards are still here and waiting for you.)
JH. |
I just found another pack of boards.
That is, I have about 25 sets that are spoken for and not paid yet (I knew that), but it also means I have another 25 sets that are still available (which I just foud out).
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
b-funk
Joined: Jul 17, 2007 Posts: 193 Location: Berlin
|
Posted: Sun Jul 27, 2008 6:49 am Post subject:
|
|
|
seems like your organisation skills are nearly as "perfect" as mine
i´m considering taking another one... |
|
Back to top
|
|
|
b-funk
Joined: Jul 17, 2007 Posts: 193 Location: Berlin
|
|
Back to top
|
|
|
Stavros
Joined: Oct 30, 2007 Posts: 36 Location: Athens-Greece
|
Posted: Sun Oct 12, 2008 11:05 am Post subject:
|
|
|
Hello, hello,
we are very quite here, any news about the PCB for the Violin/Viola/Cello Presets and the Equalizer, that shall be connected to the String Filter PCBs?
Best regards
Stavros |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Sun Oct 12, 2008 12:27 pm Post subject:
|
|
|
Stavros wrote: | Hello, hello,
we are very quite here, any news about the PCB for the Violin/Viola/Cello Presets and the Equalizer, that shall be connected to the String Filter PCBs?
Best regards
Stavros |
Hello hello -
I'm busy working on the Subtle Chorus right now.
EQs will come in 2009.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
Stavros
Joined: Oct 30, 2007 Posts: 36 Location: Athens-Greece
|
Posted: Tue Dec 09, 2008 1:14 pm Post subject:
|
|
|
Hello Jurgen,
in a previous exchange last July you have given me a response to my bellow question…
“QUOTE”
Stavros wrote:
a. Should one wish to connect the provided (40) single outs to potentiometers
Reply:
Well, it's a lot of potentiometers, and a lot of work to hook them all up, but yes, that's what the individual outs are for, if you want to go that route.
b. What would be the connection schema of one?
Reply:
Each output goes to a 1k resistor (soldered in vertically), then to a 50k log potentiometer. If you want to get rid of any DC offset, put a 1uF capacitor between the resistor and the pot.
Then sum the potentiometer wipers with 51k resistors. Either to one summing point, or to two, for stereo. You can do passive summing and go to a sensitive input on the mixing desk, or hook up some opamps.
“UNQUOTE”
Could you please help me better understand it (My electronics knowledge is not that great) elaborating a bit more on your above answer (b).
If I understand it correctly, the BPF out should be connected to the CW pin of the 50k Log Potentiometer, the CCW to Ground and the wiper to a summing point via 51k resistor. Is it correct?
Considering either a passive or active summing on one or two outputs as suggested these outputs could(?) or should(?) then be different from the normal output of the Unit. Is my assumption correct?
I will appreciate as detailed as possible clarification if necessary.
Thank you in advance for your help
Best Regards
Stavros |
|
Back to top
|
|
|
jhaible
Joined: May 25, 2007 Posts: 2014 Location: Germany
Audio files: 24
|
Posted: Tue Dec 09, 2008 2:53 pm Post subject:
|
|
|
Stavros wrote: |
If I understand it correctly, the BPF out should be connected to the CW pin of the 50k Log Potentiometer, the CCW to Ground and the wiper to a summing point via 51k resistor. Is it correct?
|
Yes, but don't connect the CW end of the 50k pot directly to the BPF output.
Put a 1k resistor directly to the BPF output (one side soldered into the PCB), and then go to th epot from the other side of the 1k resistor.
JH. _________________ "I tell you the truth, if anyone says to this mountain, 'Go, throw yourself into the sea,' and does not doubt in his heart but believes that what he says will happen, it will be done for him. Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours." (Mk 11,23f) |
|
Back to top
|
|
|
Stavros
Joined: Oct 30, 2007 Posts: 36 Location: Athens-Greece
|
Posted: Wed Dec 10, 2008 5:59 am Post subject:
|
|
|
Hi Jurgen,
thank you for your fast reply.
Could you please give me some hint also for the second part of my question?
..."Considering either a passive or active summing on one or two outputs as suggested, these outputs could(?) or should(?) then be different from the normal output of the Unit?. Is my assumption correct?"
Many thanks again
Best Regards
Stavros |
|
Back to top
|
|
|
|