OK we have some progress in the troubleshooting. There is an oscillation which is only present at the junction of pin2 and pin6 of the opamp (the two inverting inputs). It is oscillating at 1Mhz. In my early sims I did have an oscillation problem that was solved by using 40K resistors in the place where I currently have 5K resistors, so I will try upping those resistors and see what happens.
The oscilation is about +- 0.4V It is strange that it is not present on the output pins of the opamps... I don't really understand this.
unfortunately just after discovering this the wife came home and said I have to hang some curtains, so might be a while before I get to do anything about it.
Tony.
The oscilation is about +- 0.4V It is strange that it is not present on the output pins of the opamps... I don't really understand this.
unfortunately just after discovering this the wife came home and said I have to hang some curtains, so might be a while before I get to do anything about it.
Tony.
good news I think. the opa204 model that I have is very good and with the exact circuit values I have in the prototype it oscillates at somewhere a bit more than 1Mhz in the sim (at around +- 0.4V. Increasing the value of the damping resistor to 56ohms and the compensation caps from 2p to 10p seems to get rid of it. So when I get a chance I will be making those changes and hoping for clean output!
I had stopped using the opa2604 model when I started to use the opa2134, obviously the 2604 is a more accurate model!!
edit: added a couple of pics of todays earlier testing. Hopefully the next time the scope is connected there is a flat line (or close to it) when probing pins 2 and 6!
As you can see just sitting not connected to anything the scope still picks up signals in the megahertz range.
Tony.
I had stopped using the opa2604 model when I started to use the opa2134, obviously the 2604 is a more accurate model!!
edit: added a couple of pics of todays earlier testing. Hopefully the next time the scope is connected there is a flat line (or close to it) when probing pins 2 and 6!
As you can see just sitting not connected to anything the scope still picks up signals in the megahertz range.
Tony.
Attachments
Last edited:
DAC related discussion split to here --> http://www.diyaudio.com/forums/digital-line-level/257541-using-fdnrs-dac-anti-imaging-filter.html Tony.
Well I had an attempt at getting rid of the oscillation tonight. All I managed to do was blow one of the 100uF caps! I was using my two 12V batteries, but being a bit tired, and having inappropriately coloured aligator clip wires for connecting everything, I accidentally switched +12v and ground*.. oops. Lucky I was wearing my glasses, as the vapourised electrolite went straight in my face!
I had put a couple of 10pf caps piggybacked on the back of the board, but all they seem to have done is change the frequency of the oscillation
I also discovered that I DO get the oscillation on the output, I just mustn't have had the trigger adjusted to pick it up.
Also interesting was that Pin 7 has an oscillation roughly double the frequency but half the amplitude. I'm going to have to try and understand how this circuit works and what is the mechanism to cause the oscillation, or I am just going to be stabbing in the dark hoping for a solution.
I also discovered when I checked my DC offset (after replacing the cap) that it had gone from 3mV to around 70mV... I thought I must have fried something else. Took the opamp out and still had around 30mV... each B1 was contributing about 10mV... strange I checked the batteries and one was down to 10V...
The second 12V battery is now on the charger (hopefully it is not dead)... I doubt that having mismatched rails would cause the oscillation, but I guess anything is possible. I didn't actually test the batteries were properly charged on the weekend.
I'm a tenacious bugger so I will keep working on it.
Tony.
*the aligator clips are yellow for ground, black for negative and white for positive. The problem stemmed from the fact that I could only find another white aligator clip wire for connecting the positive of one battery to the negative of the other... I think you can see how I made my stupid mistake
I had put a couple of 10pf caps piggybacked on the back of the board, but all they seem to have done is change the frequency of the oscillation
I also discovered that I DO get the oscillation on the output, I just mustn't have had the trigger adjusted to pick it up. Also interesting was that Pin 7 has an oscillation roughly double the frequency but half the amplitude. I'm going to have to try and understand how this circuit works and what is the mechanism to cause the oscillation, or I am just going to be stabbing in the dark hoping for a solution.
I also discovered when I checked my DC offset (after replacing the cap) that it had gone from 3mV to around 70mV... I thought I must have fried something else. Took the opamp out and still had around 30mV... each B1 was contributing about 10mV... strange I checked the batteries and one was down to 10V...
The second 12V battery is now on the charger (hopefully it is not dead)... I doubt that having mismatched rails would cause the oscillation, but I guess anything is possible. I didn't actually test the batteries were properly charged on the weekend.
I'm a tenacious bugger so I will keep working on it.
Tony.
*the aligator clips are yellow for ground, black for negative and white for positive. The problem stemmed from the fact that I could only find another white aligator clip wire for connecting the positive of one battery to the negative of the other... I think you can see how I made my stupid mistake
Last edited:
Hmmmm, one of my "fixes" may actually be the cause of the oscillation! my 47R capacitor which is in between the bottom cap and ground, is used in this article --> ftp://ztchs.p.lodz.pl/SKUA/lab%20old/skua%20lab3.pdf to ensure oscillation! its a tad bigger, but still could be my problem.
I added it because it damped the resonances I saw in the simulation (which were doing nasty things to the phase). I guess my next test will be shorting out this resistor and seeing what the effect is!
Tony.
I added it because it damped the resonances I saw in the simulation (which were doing nasty things to the phase). I guess my next test will be shorting out this resistor and seeing what the effect is!
Tony.
Well, It would seem that the choice of opamp does make a difference to the stability of the circuit.
I tested today with fully charged batteries. dc offset was -0.8mV much better
with the opa2604 I was still getting the oscillation. It is different on each pin, I recorded all the amplitutes and frequencies.
Then I decided to try the opa2134 again. I couldn't get the oscillation, I was picking up random noise waveforms (coincidentally around the same frequency as the oscillation) which were present (and the same magnitude) whether the circuit was powered or not.
I hooked up to the sound card and measured with Holm again. MUCH better. I think that the 12pf of compensation cap has all but fixed it. There is still some fuzzyness in the knee of the response curve but it is much better, The thing that is amazingly better is the phase response.
I supsect that I could go a bit higher on the compensation, and that may clean up the remaining problems.
I'm not sure why the 2604 is oscilating wildy but the 2134 does not, but I'm not going to complain if I can get clean output with the 2134. I will probably still try the LM's but I am starting to think that it is likely to be worse (from an oscillation pov) than the 2604.
Attached is the latest holm impulse response. 2nd image shows blue with 2pf compensation, other with 12pf compensation.
Tony.
I tested today with fully charged batteries. dc offset was -0.8mV much better
with the opa2604 I was still getting the oscillation. It is different on each pin, I recorded all the amplitutes and frequencies.
Then I decided to try the opa2134 again. I couldn't get the oscillation, I was picking up random noise waveforms (coincidentally around the same frequency as the oscillation) which were present (and the same magnitude) whether the circuit was powered or not.
I hooked up to the sound card and measured with Holm again. MUCH better. I think that the 12pf of compensation cap has all but fixed it. There is still some fuzzyness in the knee of the response curve but it is much better, The thing that is amazingly better is the phase response.
I supsect that I could go a bit higher on the compensation, and that may clean up the remaining problems.
I'm not sure why the 2604 is oscilating wildy but the 2134 does not, but I'm not going to complain if I can get clean output with the 2134. I will probably still try the LM's but I am starting to think that it is likely to be worse (from an oscillation pov) than the 2604.
Attached is the latest holm impulse response. 2nd image shows blue with 2pf compensation, other with 12pf compensation.
Tony.
Attachments
Last edited:
aha, maybe the LM4562 will be better afterall. I decided to compare the specs of the opa2134 to the opa2604 (the 2134 has a lower slew rate, and lower bandwidth, but it has 20db more OLG) 104 min 120db max compared to the 2604 which is 80db min and 100db max.
The LM4562 has min 125db and typically 140db of OLG.
Tony.
The LM4562 has min 125db and typically 140db of OLG.
Tony.
hehe sorry for the excruciatingly slow progress on this James I guess the last couple of months has been a lot faster than the preceeding 4 years though!
I'm definitely feeling more confident now, and might even just live with the remaining issue, but being a perfectionist, I'm doubt I could do that, and will want to fix it.
I need to get the pots (for adjusting level of lp/hp as for the prototype I have them hardwired, so equal levels). I might put in an order today for a few LM4562's and some pots. One of the things that was stopping me doing a sound test was that I thougt I had to butcher my existing B1 preamp to do it, but that was to use the power supply. Now that I have successfully run it off two 12V batteries that's not an issue so I may give it a try on the weekend (just have to stuff the hp section.
Note that the 200Hz textbook LR filter on this test board is non-optimal for my speakers, but it will allow me to check for noise and any gross problems.
Tony.
I guess the last couple of months has been a lot faster than the preceeding 4 years though! I'm definitely feeling more confident now, and might even just live with the remaining issue, but being a perfectionist, I'm doubt I could do that, and will want to fix it.
I need to get the pots (for adjusting level of lp/hp as for the prototype I have them hardwired, so equal levels). I might put in an order today for a few LM4562's and some pots. One of the things that was stopping me doing a sound test was that I thougt I had to butcher my existing B1 preamp to do it, but that was to use the power supply. Now that I have successfully run it off two 12V batteries that's not an issue so I may give it a try on the weekend (just have to stuff the hp section.
Note that the 200Hz textbook LR filter on this test board is non-optimal for my speakers, but it will allow me to check for noise and any gross problems.
Tony.
Jeez, Tony - no complaints here - you're doing all the work and happy to see it all coming together ...
Could you send a square wave into the LP section that you're working on and see what it does with it (say 60Hz, 200, 400 and 1kHz at 1V) - that'll give a good indication if something 'got missed'
Oh, thought I'd mention this - I assume the circuit as per post #43 is still the current design altho a couple of component value changes, yes? And the numbering system as per pcb design on post #57 will be the new one?
With the change of the chip, raising the rails, increasing to the 12pF, the oscillation has been reduced dramatically, as you've said, but the freq is still about the same where the response has it's 'knee' and the phase response starts to get a 'bit radical' so I wonder if it's something outside the FDNR loop that could be causing it?
ie; looking outside the square - Perhaps the 0.1uf/2R Zobel? Or perhaps the type of the 12pF caps themselves, increasing the output buffer gate snubber, etc?
I just went over the pcb of post #57 and again, can't see anything wrong with it (hate those thru hole connections, but that's just me!)
The only point of concern is where the pads of C6 and R14 (C11 & R17 on the pcb)are very close together ...
Could you send a square wave into the LP section that you're working on and see what it does with it (say 60Hz, 200, 400 and 1kHz at 1V) - that'll give a good indication if something 'got missed'
Oh, thought I'd mention this - I assume the circuit as per post #43 is still the current design altho a couple of component value changes, yes? And the numbering system as per pcb design on post #57 will be the new one?
With the change of the chip, raising the rails, increasing to the 12pF, the oscillation has been reduced dramatically, as you've said, but the freq is still about the same where the response has it's 'knee' and the phase response starts to get a 'bit radical' so I wonder if it's something outside the FDNR loop that could be causing it?
ie; looking outside the square - Perhaps the 0.1uf/2R Zobel? Or perhaps the type of the 12pF caps themselves, increasing the output buffer gate snubber, etc?
I just went over the pcb of post #57 and again, can't see anything wrong with it (hate those thru hole connections, but that's just me!)
The only point of concern is where the pads of C6 and R14 (C11 & R17 on the pcb)are very close together ...
Hi James, just letting you know I haven't stopped on this (entirely) I did some more testing a week or two ago and I'm pretty sure that with the opa2134 I have the oscillation problem licked.
I also found that the phase rising above about 5K was the sound card, when running balanced to SE and back to balanced.. I did a loopback with my balanced to SE / SE to balanced cables and found that it rose significantly. I was unable to do a calibration in holm because it kept crashing, so I tried uninstalling the 64bit version and installing the 32bit version instead, and that let me do a calibration. The result is much better looking phase
Can't attach a picture right now because I'm on the laptop and the measurements were done on my desktop at home.
Have had some other diy (not electronics) projects mandated by the other half lately so it is on hold a bit for now.
Tony.
I did some more testing a week or two ago and I'm pretty sure that with the opa2134 I have the oscillation problem licked. I also found that the phase rising above about 5K was the sound card, when running balanced to SE and back to balanced.. I did a loopback with my balanced to SE / SE to balanced cables and found that it rose significantly. I was unable to do a calibration in holm because it kept crashing, so I tried uninstalling the 64bit version and installing the 32bit version instead, and that let me do a calibration. The result is much better looking phase
Can't attach a picture right now because I'm on the laptop and the measurements were done on my desktop at home.
Have had some other diy (not electronics) projects mandated by the other half lately so it is on hold a bit for now.
Tony.
No problems, Tony - these things take time, and we're getting towards springtime ....
That's really good that you've found the source of the increasing phase response as it has/had all the hallmarks of a stability problem - can that program of yours do the old "bode plots thing" as I vaguely remember that this was instrumental in locating pole/zeros that could be related back to finding phase instability sources - it's been a whole lot of years since I've looked at this sort of thing, possibly the 'holm, etc" can do a better job.
I'm having a losing battle with one of those damn programs - - trying to combine the 'standard practice' deflectors/diffusors/absorbtion with lamina flow, fluid mechanics, etc in the rear of the bass chambers - talk about contradictions - I think ever Floyd Toole would get indigestion!
It started off as a 'sort-of' Sudoki exercise (like designing pcbs) ....!
That's really good that you've found the source of the increasing phase response as it has/had all the hallmarks of a stability problem - can that program of yours do the old "bode plots thing" as I vaguely remember that this was instrumental in locating pole/zeros that could be related back to finding phase instability sources - it's been a whole lot of years since I've looked at this sort of thing, possibly the 'holm, etc" can do a better job.
I'm having a losing battle with one of those damn programs - - trying to combine the 'standard practice' deflectors/diffusors/absorbtion with lamina flow, fluid mechanics, etc in the rear of the bass chambers - talk about contradictions - I think ever Floyd Toole would get indigestion!
It started off as a 'sort-of' Sudoki exercise (like designing pcbs) ....!
Hi James, hope you have made some progress, sounds complex!!
I thought I'd post an update, I hadn't been happy with my measurements that I was trying to use to come up with my low pass and high pass filters, so I decided to try some more in room measurements now that I have curtains on the back wall.
Major improvement. The phase was a bit all over the place, but 1 octave smoothing fixed that and gave me something to work with in speaker workshop. I'm hoping this is a valid path, I guess I'll know once I construct the filters.
Attached are the holm in room measurements for my MTM's and the old 3ways (only the 10" vifa's running). first 1/8th octave smoothing, then 1 octave smoothing. The impulse on the woofers indicates that my thoughts of the need for a cabinet redesign (for better transient response) are justified.
Then the Speaker workshop sim results. 3rd order on each woofer and MTM, nice flat result and excellent reverse null.
Pretty happy with this! Sims ok in spice as well.
The speaker workshop sim is done with a modified impedance file for each of the woofer and MTM. I took the actual impedance measurements loaded them into excell and changed the impedance colum to be flat. 6.5K for the High pass (MTM) and 1 ohm for the lowpass (woofer).
6.5K is the value that gets used in the actual circuit for the high pass, 1 ohm is the normalized value for the low pass filter (which translates to a 0.1 uF cap).
I'll be very interested to see how this turns out.
edit2: I've also attached speaker workshop sim of stock 3rd order at 250 Hz (the same freqency that the first sim is centred on). This is why I want to design the crossover this "unconventional" way!! The difference is quite large! I think that this in room approach is best as it is allowing me to correct for some room response abhoration as well
Tony.
I thought I'd post an update, I hadn't been happy with my measurements that I was trying to use to come up with my low pass and high pass filters, so I decided to try some more in room measurements now that I have curtains on the back wall.
Major improvement. The phase was a bit all over the place, but 1 octave smoothing fixed that and gave me something to work with in speaker workshop. I'm hoping this is a valid path, I guess I'll know once I construct the filters.
Attached are the holm in room measurements for my MTM's and the old 3ways (only the 10" vifa's running). first 1/8th octave smoothing, then 1 octave smoothing. The impulse on the woofers indicates that my thoughts of the need for a cabinet redesign (for better transient response) are justified.
Then the Speaker workshop sim results. 3rd order on each woofer and MTM, nice flat result and excellent reverse null.
Pretty happy with this! Sims ok in spice as well.
The speaker workshop sim is done with a modified impedance file for each of the woofer and MTM. I took the actual impedance measurements loaded them into excell and changed the impedance colum to be flat. 6.5K for the High pass (MTM) and 1 ohm for the lowpass (woofer).
6.5K is the value that gets used in the actual circuit for the high pass, 1 ohm is the normalized value for the low pass filter (which translates to a 0.1 uF cap).
I'll be very interested to see how this turns out.
edit2: I've also attached speaker workshop sim of stock 3rd order at 250 Hz (the same freqency that the first sim is centred on). This is why I want to design the crossover this "unconventional" way!! The difference is quite large! I think that this in room approach is best as it is allowing me to correct for some room response abhoration as well
Tony.
Attachments
Last edited:
The first plot is your 'in room' active response? The freq response appears quite reasonable with about a +/- 5dB range - I assume the phase response is the dotted line and is just showing the effects of room nodes standing waves, I think.
Not quite sure what the Impulse response is saying - is the blue line the reflection off the back wall? If so, it's a fairly quiet room if done during daytime - you'll get a nice surprise when you measure the effects of a Schroedder diffuser on the rear wall
On the second plot, this is the 'smoothed', yes? Easier to read - what's happening at 500Hz to cause this dramatic phase shift? one of the drivers connected back to front? (green dotted line?) The freq response appears to be pretty good so maybe this is just another anomaly ....
Graph 3 seems too good to be true - this is the SIM? [Ahh, I see the Xover at 300Hz]
Graph 4 is the 2 responses done separately? Wish I could get this sort of result in practice!
Not much progress in the bass chamber diffusers - gave the bass driver a good 'kick in the A' today with about 25 volts signal from Greg Ball's gb150D amp - lightweight panels are new venture for me and appear to produce lots of rattles/noises - time to start gluing on the leaded vinyl and Acoustimas, I think!
At least it certainly has enormous 'kick' but it's going to need a ridgid front mounting panel - maybe one of those concrete/composite ones - even with the unfinished box/baffles, I thought the bloody windows were going to break!
I had a quick 'squiz' at those PLL filters that a lot of guys are quite enamoured about - know anything about them?
Not quite sure what the Impulse response is saying - is the blue line the reflection off the back wall? If so, it's a fairly quiet room if done during daytime - you'll get a nice surprise when you measure the effects of a Schroedder diffuser on the rear wall
On the second plot, this is the 'smoothed', yes? Easier to read - what's happening at 500Hz to cause this dramatic phase shift? one of the drivers connected back to front? (green dotted line?) The freq response appears to be pretty good so maybe this is just another anomaly ....
Graph 3 seems too good to be true - this is the SIM? [Ahh, I see the Xover at 300Hz]
Graph 4 is the 2 responses done separately? Wish I could get this sort of result in practice!
Not much progress in the bass chamber diffusers - gave the bass driver a good 'kick in the A' today with about 25 volts signal from Greg Ball's gb150D amp - lightweight panels are new venture for me and appear to produce lots of rattles/noises - time to start gluing on the leaded vinyl and Acoustimas, I think!
At least it certainly has enormous 'kick' but it's going to need a ridgid front mounting panel - maybe one of those concrete/composite ones - even with the unfinished box/baffles, I thought the bloody windows were going to break!
I had a quick 'squiz' at those PLL filters that a lot of guys are quite enamoured about - know anything about them?
Hi James, first plot and second are the same just the smoothing is different, yes in room response, and dotted lines are the phase. Blue is the 10" running full range (no crossover) and green is the MTM running as normal (no crossover than it's own passive).
Blue line in the impulse response is the woofers response, green line is the MTM's response. I'm thinking that the impulse response looks like the sort of step response that unibox gives when you have a flabby reflex tuning.
Yes graph 3 is the sim, but based on my passive sims with speaker workshop it should be pretty accurate (assuming this approach works)... It is 1 octave smoothed so real in room will be nothing near as smooth, but it should hopefully intergrate well.
graph 4 is same as graph 3 with MTM reversed in polarity.
Note the measurements were taken at the listening position with both left and right speakers, I possibly should have taken a few measurememts perhaps + - 0.5M iether side of the listening position and averaged them, but I wasn't sure how that would go for getting the phase relationship between the woofers and MTM's.
One thing I want to try when I make my new bass cabinets is constrained layer. I saw something on TV ages ago where they had a thin rubber film that was used between gyprock sheets which was amazingly effective at deadening. I want to try the same with a 9mm marine ply and 12mm MDF sandwich. Only problem is I don't know what the stuff was called.
Tony.
Blue line in the impulse response is the woofers response, green line is the MTM's response. I'm thinking that the impulse response looks like the sort of step response that unibox gives when you have a flabby reflex tuning.
Yes graph 3 is the sim, but based on my passive sims with speaker workshop it should be pretty accurate (assuming this approach works)... It is 1 octave smoothed so real in room will be nothing near as smooth, but it should hopefully intergrate well.
graph 4 is same as graph 3 with MTM reversed in polarity.
Note the measurements were taken at the listening position with both left and right speakers, I possibly should have taken a few measurememts perhaps + - 0.5M iether side of the listening position and averaged them, but I wasn't sure how that would go for getting the phase relationship between the woofers and MTM's.
One thing I want to try when I make my new bass cabinets is constrained layer. I saw something on TV ages ago where they had a thin rubber film that was used between gyprock sheets which was amazingly effective at deadening. I want to try the same with a 9mm marine ply and 12mm MDF sandwich. Only problem is I don't know what the stuff was called.
Tony.
Have a look at that Acoustiflex-Green - it's made by the Acoustica people up in Annandale NSW (acoustica.com.au) - comes in a roll - suggest you avoid MDF and use Aussie ply, not Baltic
For your room, have a look at the range of panels from 'Vicoustic' - retail agency in South Aus, I think - far cheaper and just as effective as the well known RPG. etc products
Also, a guy called Jim Smith has written a book and DVD about setting up rooms properly - " Get Better Sound" - an absolute gem - simple explanations and easy-to-do setups for every audio/diy person (and every sales/demo room should use it too!)
For your room, have a look at the range of panels from 'Vicoustic' - retail agency in South Aus, I think - far cheaper and just as effective as the well known RPG. etc products
Also, a guy called Jim Smith has written a book and DVD about setting up rooms properly - " Get Better Sound" - an absolute gem - simple explanations and easy-to-do setups for every audio/diy person (and every sales/demo room should use it too!)
One of the things with simulated inductors is that they don't 100% match what the theory says they will do atfer getting the values in speakerworkshop and translating them to the nominal values for the filter I simulate the passive network as well as the active one. There is usually some deviation. after a bit of tweaking I got a pretty good match for both using standard values.
Attached is the spice sim showing the active against the passive as you can see pretty close. You can also see the deviation from a standard rolloff with the HP filter, but this is the desired effect. The filter slope combined with the actual speakers response in room should give close to a perfect 4th order Bessel slope at 250Hz.
I went and bought the resistors today, that I need to implement this version. Won't get a chance tonight but perhaps tomorrow
the second image shows the circuit with passive as per the speaker workshop sim (with perfect coils and caps) and the active tuned to match. You can see there are a few differences. In the high pass the load resistor was increased 1K and the "inductor" resistor reduced from 3.65k to 3.6k.
Low pass almost everything was adjusted to get the curve to match! Note that this version of the schematic is well and truly customized to my particular speakers in my room! If I used this version in another room it would most likely be problematic.
Tony.
Attached is the spice sim showing the active against the passive as you can see pretty close. You can also see the deviation from a standard rolloff with the HP filter, but this is the desired effect. The filter slope combined with the actual speakers response in room should give close to a perfect 4th order Bessel slope at 250Hz.
I went and bought the resistors today, that I need to implement this version. Won't get a chance tonight but perhaps tomorrow
the second image shows the circuit with passive as per the speaker workshop sim (with perfect coils and caps) and the active tuned to match. You can see there are a few differences. In the high pass the load resistor was increased 1K and the "inductor" resistor reduced from 3.65k to 3.6k.
Low pass almost everything was adjusted to get the curve to match! Note that this version of the schematic is well and truly customized to my particular speakers in my room!
If I used this version in another room it would most likely be problematic. Tony.
Attachments
Oh, my!
It is one thing to compensate the filters for anomalies in your drivers performance (freq response) but trying to do the same for room acoustics is something altogether different (time domain) - I'd suggest you initially just sort out the filters for the freq response and see how it turns out, and then maybe look at some room treatment &/or some limited freq compensation, but that's just me ....
Is that your passive Xover in the bottom left corner with the 800uF, 1.75mH, 0.3mH filter and the high pass one below it? Doesn't your MTM require some compensation/baffle step filters - and weird values ...
It is one thing to compensate the filters for anomalies in your drivers performance (freq response) but trying to do the same for room acoustics is something altogether different (time domain) - I'd suggest you initially just sort out the filters for the freq response and see how it turns out, and then maybe look at some room treatment &/or some limited freq compensation, but that's just me ....
Is that your passive Xover in the bottom left corner with the 800uF, 1.75mH, 0.3mH filter and the high pass one below it? Doesn't your MTM require some compensation/baffle step filters - and weird values ...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.