Hi To All,
I'm doing at some research in advance of purchasing a microphone to begin taking my own speaker measurements. My questions are centered around what many measurements do I need to capture for Xover design in PCD7. I would assume that the measurements required would be as follows:
1. Tweeter by itself (farfield)
2. Midrange by itself (farfield)
3. Top Woofer by itself (farfield)
4. Bottom Woofer by itself (farfield)
5. Tweeter with midrange (farfield)
6. Midrange with both woofers (farfield)?
7. Tweeter with both woofers (farfield)?
8. Both woofers driven together (farfield)
9. Nearfield of 1 woofer for merging with farfield measurement
10. Nearfield of port
Is my assumptions of required measurements correct or have I missed anything? Obviously, this will be new territory for me. Up to this point, I done builds based on simulations with an eye towards someday tweaking the designs based on actual measurements.
Best Regards,
Rich
I'm doing at some research in advance of purchasing a microphone to begin taking my own speaker measurements. My questions are centered around what many measurements do I need to capture for Xover design in PCD7. I would assume that the measurements required would be as follows:
1. Tweeter by itself (farfield)
2. Midrange by itself (farfield)
3. Top Woofer by itself (farfield)
4. Bottom Woofer by itself (farfield)
5. Tweeter with midrange (farfield)
6. Midrange with both woofers (farfield)?
7. Tweeter with both woofers (farfield)?
8. Both woofers driven together (farfield)
9. Nearfield of 1 woofer for merging with farfield measurement
10. Nearfield of port
Is my assumptions of required measurements correct or have I missed anything? Obviously, this will be new territory for me. Up to this point, I done builds based on simulations with an eye towards someday tweaking the designs based on actual measurements.
Best Regards,
Rich
Attachments
If you are doing quasi-anechoic, you want the near-field individual driver responses, plus acoustic distances. Then add baffle step correction.
The far-field for the mid/woofer will help ensure your baffle step adjustment is in fact accurate.
For distances, it is OK to measure this additively. That is, measure tweeter to mid, and mid to woofers (keeping the mic in a single fixed position for each, at least 3'/1m away). If you measure:
Tweet to Mid: 1"
Mid to woofer: 2"
Then:
Tweet: 0"
Mid: 1"
Woofer: 3"
With multiple drivers, identically driven, I cheat.
I measure them together (including the impedance) but then I adjust the FRD and ZMA files so I can model them individually. Exactly how depends on how you measured them. On the other hand, the simpler method is to measure both woofers together, and treat as a single driver in your simulation tool.
Best,
E
The far-field for the mid/woofer will help ensure your baffle step adjustment is in fact accurate.
For distances, it is OK to measure this additively. That is, measure tweeter to mid, and mid to woofers (keeping the mic in a single fixed position for each, at least 3'/1m away). If you measure:
Tweet to Mid: 1"
Mid to woofer: 2"
Then:
Tweet: 0"
Mid: 1"
Woofer: 3"
With multiple drivers, identically driven, I cheat.
I measure them together (including the impedance) but then I adjust the FRD and ZMA files so I can model them individually. Exactly how depends on how you measured them. On the other hand, the simpler method is to measure both woofers together, and treat as a single driver in your simulation tool. Best,
E
Last edited:
Hi Erik,
I really appreciate your input! Could you explain the logic of doing nearfield measurements of tweeter and midrange? Not doubting your knowledge, just wanting to also learn the Why's along the journey. I have access to a very large room for taking measurements. My goal is to try and capture FR down to 200 hz or so. That might be a lofty goal, but I will soon find out. Also, since most Xover design work will be in PCD7; I would like to follow Jeff Bagby's detailed write ups ' Accurate In-Room Response To 10 HZ' and 'Finding The Relative Acoustic Offset In PCD'. Everything he mentions is understandable (even for a noob) but his examples describe a 2 way design. I'm sure the principles remain the same, but there will need to be additional merging of measurements. Which, will lead to further questions at the next turn.
Best,
Rich
I really appreciate your input! Could you explain the logic of doing nearfield measurements of tweeter and midrange? Not doubting your knowledge, just wanting to also learn the Why's along the journey.
I have access to a very large room for taking measurements. My goal is to try and capture FR down to 200 hz or so. That might be a lofty goal, but I will soon find out. Also, since most Xover design work will be in PCD7; I would like to follow Jeff Bagby's detailed write ups ' Accurate In-Room Response To 10 HZ' and 'Finding The Relative Acoustic Offset In PCD'. Everything he mentions is understandable (even for a noob) but his examples describe a 2 way design. I'm sure the principles remain the same, but there will need to be additional merging of measurements. Which, will lead to further questions at the next turn. Best,
Rich
Personally, I make 2-way systems, so I can ROYALLY cheat and do pretty much everything with farfield, adjusting for my target curve in place.
However, this method only works so long as I can use gated measurements through the crossover section. Below this point, boundary interference becomes so severe that measurements are at risk of being inaccurate. Typically, this means you can only do this cheat with a 2-way system, and below that, you are forced to use near-field measurements. This is referred to as quasi-anechoic.
Meaning, we use near-field to increase the direct to reflected SPL to the point that the room becomes irrelevant. D'Appolito covers this in a great depth. And while he is accurate, more recent work (by Jeff Bagby?) has shown that the near-field alone doesn't take into account the uneven reinforcement of the front baffle. That is, alone, D'Appolito's work (and that of his contemporaries) was incomplete.
So, yeah, you hit on something. I use far field to guide my near-field splicing. OmniMic really really helps you do this very well. Also, far field will help you see the overall bass response with high credibility. You'll have a lot of noise from room modes, but it's the best way to know you aren't lying to yourself when you finish splicing everything together.
Best,
E
However, this method only works so long as I can use gated measurements through the crossover section. Below this point, boundary interference becomes so severe that measurements are at risk of being inaccurate. Typically, this means you can only do this cheat with a 2-way system, and below that, you are forced to use near-field measurements. This is referred to as quasi-anechoic.
Meaning, we use near-field to increase the direct to reflected SPL to the point that the room becomes irrelevant. D'Appolito covers this in a great depth. And while he is accurate, more recent work (by Jeff Bagby?) has shown that the near-field alone doesn't take into account the uneven reinforcement of the front baffle. That is, alone, D'Appolito's work (and that of his contemporaries) was incomplete.
So, yeah, you hit on something. I use far field to guide my near-field splicing. OmniMic really really helps you do this very well. Also, far field will help you see the overall bass response with high credibility. You'll have a lot of noise from room modes, but it's the best way to know you aren't lying to yourself when you finish splicing everything together.
Best,
E
See here:
How to use OmniMic and PCD to find the Relative Acoustic Offset - Techtalk Speaker Building, Audio, Video Discussion Forum
or better the linked PDF document: https://app.box.com/s/ouxjjsx0m8bs00cil5iq
and also this document: http://audio.claub.net/software/FRD_Blender/White%20Paper%20-%20Accurate%20In-Room%20Frequency%20Response%20to%2010Hz.pdf
From that it is evident that you need tweeter, midrange and tweeter+midrange far field, possibly midrange near field, and woofer far and near field + tube response. For the woofer the measurement should be performed with both attached, the near field will be performed to only one woofer.
Hope this helps.
Ralf
How to use OmniMic and PCD to find the Relative Acoustic Offset - Techtalk Speaker Building, Audio, Video Discussion Forum
or better the linked PDF document: https://app.box.com/s/ouxjjsx0m8bs00cil5iq
and also this document: http://audio.claub.net/software/FRD_Blender/White%20Paper%20-%20Accurate%20In-Room%20Frequency%20Response%20to%2010Hz.pdf
From that it is evident that you need tweeter, midrange and tweeter+midrange far field, possibly midrange near field, and woofer far and near field + tube response. For the woofer the measurement should be performed with both attached, the near field will be performed to only one woofer.
Hope this helps.
Ralf
Hi Ralf,
Thanks so much for your reply! The links you posted are the exact methods I plan going to follow. I was pointed towards Jeff Bagby's excellent programs when I first got into this hobby a couple years ago. So, it makes since to me to follow his detailed write-ups moving forward. Your comment on driving both woofers at the same time helps me out a bunch. That is the one area where I was unsure of the correct measurement method in dual woofer designs.
Best Regards,
Rich
Thanks so much for your reply! The links you posted are the exact methods I plan going to follow. I was pointed towards Jeff Bagby's excellent programs when I first got into this hobby a couple years ago. So, it makes since to me to follow his detailed write-ups moving forward. Your comment on driving both woofers at the same time helps me out a bunch.
That is the one area where I was unsure of the correct measurement method in dual woofer designs. Best Regards,
Rich
Thanks Erik,
I will keep that in mind. PCD7 is a little different then Xsim in that you assign baffle coordinates for both woofers (i.e x,y, and z offsets). The challenge for me when I get to that point is the order in which files are imported into PCD to dial in the correct offsets. So, in other words I still have much to learn and can use all the help I can get.
Best,
Rich
I will keep that in mind. PCD7 is a little different then Xsim in that you assign baffle coordinates for both woofers (i.e x,y, and z offsets). The challenge for me when I get to that point is the order in which files are imported into PCD to dial in the correct offsets. So, in other words I still have much to learn and can use all the help I can get.
Best,
Rich
Speaker workshop is pretty handy for scaling nearfield / farfield. I presume you're familiar with Bagby and Laub's FRD blender tool?:
FRD Blender and Minimum Phase Extractor
My process - usually 1m distance is enough. I do outside measurements with tweeter at least 2.4m in the air to get 1,024 sample points with a >11msec gate at 96KHz. I use speaker workshop for measurements, but FRD tools to extract minimum phase since any "munging" of curves means recorded phase in SW is meaningless. Driver offsets must be factored then in the xover sim as Erik pointed out.
1. Gated tweeter (farfield)
2. Gated mid (farfield)
3. Gated woofer (farfield)
4. Nearfield mid
5. Nearfield mid port (if one)
6. Merge mid nearfield and port together (in SW)
7. Nearfield woofer
8. Nearfield woofer port (if one)
9. Merge woofer nearfield and port together (in SW)
10. Scale mid merged curve to mid farfield response
11. Scale woofer merged curve to woofer farfield response
12. Run FRD blender to splice merged nearfield into farfield for mid - check levels. Extract min phase
13. Run FRD blender to splice merged nearfield into farfield for woofer - check levels. Extract min phase
14. Extract min phase for tweeter
15. Re-import all into SW
16. have some xover fun.
FRD Blender and Minimum Phase Extractor
My process - usually 1m distance is enough. I do outside measurements with tweeter at least 2.4m in the air to get 1,024 sample points with a >11msec gate at 96KHz. I use speaker workshop for measurements, but FRD tools to extract minimum phase since any "munging" of curves means recorded phase in SW is meaningless. Driver offsets must be factored then in the xover sim as Erik pointed out.
1. Gated tweeter (farfield)
2. Gated mid (farfield)
3. Gated woofer (farfield)
4. Nearfield mid
5. Nearfield mid port (if one)
6. Merge mid nearfield and port together (in SW)
7. Nearfield woofer
8. Nearfield woofer port (if one)
9. Merge woofer nearfield and port together (in SW)
10. Scale mid merged curve to mid farfield response
11. Scale woofer merged curve to woofer farfield response
12. Run FRD blender to splice merged nearfield into farfield for mid - check levels. Extract min phase
13. Run FRD blender to splice merged nearfield into farfield for woofer - check levels. Extract min phase
14. Extract min phase for tweeter
15. Re-import all into SW
16. have some xover fun.
Thanks Dave,
Really appreciate you listing the steps required from start to finish. I'm very familiar with FRD Blender Tool. I will be working with Room EQ Wizard along with Umik-1 mic but I'm sure the flow of work is the same or very close. I am confident of one thing only and that is I'm sure to encounter snags along the way. That's where it is helpful to rely on the advice and direction from members of this forum.
Best Regards,
Rich
Really appreciate you listing the steps required from start to finish. I'm very familiar with FRD Blender Tool. I will be working with Room EQ Wizard along with Umik-1 mic but I'm sure the flow of work is the same or very close. I am confident of one thing only and that is I'm sure to encounter snags along the way.
That's where it is helpful to rely on the advice and direction from members of this forum.Best Regards,
Rich
This was originally written for measuring closed WWMTMWW speaker with ARTA and processing with VituixCAD: VituixCAD Measurement Preparations.pdf. Vented system needs one measurement for each port.
Simulator is here: VituixCAD.
This method and simulator might be too professional and comprehensive in the beginning (also for experienced members on diyaudio). Another problem is that hobby might turn dull because good results are so easy to achieve without endless subjective tuning.
Simulator is here: VituixCAD.
This method and simulator might be too professional and comprehensive in the beginning (also for experienced members on diyaudio). Another problem is that hobby might turn dull because good results are so easy to achieve without endless subjective tuning.
^It is not easier than a year ago, though new features such as crossover optimizer help you achieve your goals faster. Video lessons should help. For example Optimizer lesson shows how crossover designing goes.
I think that most of the diy speakers need so much work and crossover parts are expensive that projects deserve dual channel (with timing reference) polar response measurements and crossover simulator which can handle also directivity and power response to full or half space. Good thing is that capable tool is available for free. Situation was totally different only few ears ago.
I think that most of the diy speakers need so much work and crossover parts are expensive that projects deserve dual channel (with timing reference) polar response measurements and crossover simulator which can handle also directivity and power response to full or half space. Good thing is that capable tool is available for free. Situation was totally different only few ears ago.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.