You should be able to see imported measurements immediately. Can you provide the files that you are using? Do those files use a comma for the decimal separator? WinPCD requires a period. Jeff's PCD may have been able to handle a comma given that it's in Excel and you probably have a Euro version of Excel (comma for decimal separator, correct?), but if it's a CSV file with comma separator, I didn't provide a way to import those, it requires the period. It will ignore lines with the extra commas. I should add that to my to-do list.
Dave
Dave
Hello.
I checked the file, and all of them come with a period as decimal separator. Here attached you'll find tho of them.
I installed the last win 10 updates to be sure i've no problems on that side. Framework is updated to the last 4.8 rev.
I checked the file, and all of them come with a period as decimal separator. Here attached you'll find tho of them.
I installed the last win 10 updates to be sure i've no problems on that side. Framework is updated to the last 4.8 rev.
All files load in the latest version, 1.612. All three frd files look fine, although all files have no phase at all. But the woofer and tweeter impedance files don't look right at all. The magnitude is so high that they are above the default range of the graph. If you use the slider on the side to move those down you'll see them. But they can't be correct for raw driver file impedance, they are way too high for magnitude. They all have no phase, either.
To design, you will have to generate phase. You can use my WinPhase app for the frd files, but I didn't write it to work for zma. Looks like I need to create one for impedance files since they'll need slightly different ways to set up the magnitude curve. There are other apps or Excel spreadsheets (maybe one of Jeff's) to do that. I've never looked into that, I always have measured data files.
When you generate phase for the tweeter file (TN101.frd) use 210Hz for the highpass, that will eliminate the bad data below that point. I think that the (midrange?) file 10fe330 FRD-png.frd should have highpass at about 27Hz. The woofer file WSN041.frd may be best at 60Hz highpass, but the default is very close, so for crossover design either one will do.
For the frd lowpass in all three you have no measured phase to try to approximate that when generating phase, so you'll probably be best using the defaults.
Again, you'll need to find a way to generate phase for the impedance files. I add an app to my to-do list for magnitude, but you need to find one now. Check Jeff Bagby's Excel spreadsheets since you have Excel, I think one of his will work for you.
Dave
To design, you will have to generate phase. You can use my WinPhase app for the frd files, but I didn't write it to work for zma. Looks like I need to create one for impedance files since they'll need slightly different ways to set up the magnitude curve. There are other apps or Excel spreadsheets (maybe one of Jeff's) to do that. I've never looked into that, I always have measured data files.
When you generate phase for the tweeter file (TN101.frd) use 210Hz for the highpass, that will eliminate the bad data below that point. I think that the (midrange?) file 10fe330 FRD-png.frd should have highpass at about 27Hz. The woofer file WSN041.frd may be best at 60Hz highpass, but the default is very close, so for crossover design either one will do.
For the frd lowpass in all three you have no measured phase to try to approximate that when generating phase, so you'll probably be best using the defaults.
Again, you'll need to find a way to generate phase for the impedance files. I add an app to my to-do list for magnitude, but you need to find one now. Check Jeff Bagby's Excel spreadsheets since you have Excel, I think one of his will work for you.
Dave
Good Morning.
Thanks so much for your help and suggestion.
Meanwhile i was trying to use one tool from Vituixcad to create Frd and ZMA files.
The tool create just txt file..do you think that renaming those file would be enough..
Here one i created so far.
Thanks again.
Maurizio
Thanks so much for your help and suggestion.
Meanwhile i was trying to use one tool from Vituixcad to create Frd and ZMA files.
The tool create just txt file..do you think that renaming those file would be enough..
Here one i created so far.
Thanks again.
Maurizio
I made a check with the files generated with Vituix Cad..and following your suggestions as well...but not graphs..
I was trying moving the slider...but nothing there..
Should i change setting on my windows?
I will try to get some frd files from other sources to make a cross check.
I was trying moving the slider...but nothing there..
Should i change setting on my windows?
I will try to get some frd files from other sources to make a cross check.
The filename extension can actually be anything but the program open dialog by default looks for either frd or txt. Both are actually the same kind of file, a text file, only the extension name is different. What matters is what kind of file it is. You can rename a txt to frd, but it's not required.
I've never seen a control on a form disappear like that for a the tweeter or any other control that isn't specifically coded to do that. When you change from a 3-way to a 2-way the midrange controls are hidden on the System tab, but never the tweeter. The code never changes any tweeter control on the main form so I have no idea how that would have disappeared. If you can supply the files for this test you made maybe I can see what might do that, but I know that it's not built into the code. If there was some kind of bug in the forms code, it would help to see what the Tweeter tab looks like. But I can't do anything without the files you imported.
Dave
I've never seen a control on a form disappear like that for a the tweeter or any other control that isn't specifically coded to do that. When you change from a 3-way to a 2-way the midrange controls are hidden on the System tab, but never the tweeter. The code never changes any tweeter control on the main form so I have no idea how that would have disappeared. If you can supply the files for this test you made maybe I can see what might do that, but I know that it's not built into the code. If there was some kind of bug in the forms code, it would help to see what the Tweeter tab looks like. But I can't do anything without the files you imported.
Dave
'afternoon, Dave. Long-time WinPCD-user here, alongside Jeff's Excel sheet, & really enjoying the most recent updates / enhancements you've made.
Can I be an extremely cheeky gibbon & make a suggestion / request? Above all because you'll probably take one look & think a/ I've taken leave of my senses, and b/ it's a whole lot of work. But I'll ask anyway. 😉
Since WinPCD (and PCD) aren't free-form, and realistically speaking can't be, there are some limits on what topologies are available. You've already removed one of those by allowing a tank notch over the primary series inductor / cap in a parallel circuit, which is something PCD can't handle -many thanks for that. As an extension: would there be any way you could alter the shunt elements in the main filter section (e.g. 2nd, 3rd, 4th order) to a full LCR, rather than 'just' a CR or LR depending on whether it's a low or high-pass? I know there are the LCR shunts after the nominal main filter elements, & most of the time this is okay, but it doesn't always give enough freedom. For e.g.: say you wanted to do a modified 3rd order HP, with an LCR as the shunt element, followed by an L-pad and an Fs LCR Zobel (I'm not saying that's something I'm trying -it's just an example). At the moment, it can't be done, the available topology / ladder doesn't allow it. But if the shunt elements of the main filter were adapted to full LCR status: more complex circuit topologies become possible, even though it's not a complete free-form circuit. Looking at the data entry tabs for HP, mid & LP, there's physically space to do it without needing to increase the screen size, since there are already three potential data entry points for the series elements, but only two for the shunt.
Cheers (and apologies for the major PITA question 😉 )
Scott
Can I be an extremely cheeky gibbon & make a suggestion / request? Above all because you'll probably take one look & think a/ I've taken leave of my senses, and b/ it's a whole lot of work. But I'll ask anyway. 😉
Since WinPCD (and PCD) aren't free-form, and realistically speaking can't be, there are some limits on what topologies are available. You've already removed one of those by allowing a tank notch over the primary series inductor / cap in a parallel circuit, which is something PCD can't handle -many thanks for that. As an extension: would there be any way you could alter the shunt elements in the main filter section (e.g. 2nd, 3rd, 4th order) to a full LCR, rather than 'just' a CR or LR depending on whether it's a low or high-pass? I know there are the LCR shunts after the nominal main filter elements, & most of the time this is okay, but it doesn't always give enough freedom. For e.g.: say you wanted to do a modified 3rd order HP, with an LCR as the shunt element, followed by an L-pad and an Fs LCR Zobel (I'm not saying that's something I'm trying -it's just an example). At the moment, it can't be done, the available topology / ladder doesn't allow it. But if the shunt elements of the main filter were adapted to full LCR status: more complex circuit topologies become possible, even though it's not a complete free-form circuit. Looking at the data entry tabs for HP, mid & LP, there's physically space to do it without needing to increase the screen size, since there are already three potential data entry points for the series elements, but only two for the shunt.
Cheers (and apologies for the major PITA question 😉 )
Scott
Hi Scott,
Always good to hear from users. I haven't been working on any changes since the last release. What you ask is doable, it doesn't require a big change for the matrix calculations, just inserting the new elements. The change to the circuit overlay takes a bit more, since it's a fixed graphic to start, but it certainly can be made. It may be confusing to some users, especially new ones seeing the added complication in the layout (not a frequent use I would think), but I'll put it in a to-do list. I'm spending all my time now on a near-field/distortion measurement rig (aside from other activities), so I'll probably not get to that for some time. Too much to do, not enough time.
Good suggestion, though. I have another addition I'd like to make for myself, but that one requires quite a bit of change of the matrix layout for calculations, so it's been on hold for a long time now. I may look into that at the same time.
Dave
I do need to think about how to do this. When a user selects a XO topology, only those typical for that topology appear. The full LCR set shouldn't be enabled, only the proper LR/CR should. The extra element isn't part of a typical topology. Likewise, when the initialization for "textbook" values is brought up, it won't include any of those, so it's going to make it more confusing. I may need to make those individually enable/disabled in a different way. My goal has always been to make it as user-friendly as possible.
Always good to hear from users. I haven't been working on any changes since the last release. What you ask is doable, it doesn't require a big change for the matrix calculations, just inserting the new elements. The change to the circuit overlay takes a bit more, since it's a fixed graphic to start, but it certainly can be made. It may be confusing to some users, especially new ones seeing the added complication in the layout (not a frequent use I would think), but I'll put it in a to-do list. I'm spending all my time now on a near-field/distortion measurement rig (aside from other activities), so I'll probably not get to that for some time. Too much to do, not enough time.
Good suggestion, though. I have another addition I'd like to make for myself, but that one requires quite a bit of change of the matrix layout for calculations, so it's been on hold for a long time now. I may look into that at the same time.
Dave
I do need to think about how to do this. When a user selects a XO topology, only those typical for that topology appear. The full LCR set shouldn't be enabled, only the proper LR/CR should. The extra element isn't part of a typical topology. Likewise, when the initialization for "textbook" values is brought up, it won't include any of those, so it's going to make it more confusing. I may need to make those individually enable/disabled in a different way. My goal has always been to make it as user-friendly as possible.
Last edited:
Hello Dave.
I found the issue, and was related to Windows. As i changed the decimal separator from comma to period it started to work.
So finally the graphs are not empty anymore.
I could upload the txt files from Vituix Cad as you already told me.
Here attached two of them..can you check if these files are ok?
By the way...your SW is really great. I'm not expert with crossovers and is really helpful.
Cheers from Italy.
Maurizio
I found the issue, and was related to Windows. As i changed the decimal separator from comma to period it started to work.
So finally the graphs are not empty anymore.
I could upload the txt files from Vituix Cad as you already told me.
Here attached two of them..can you check if these files are ok?
By the way...your SW is really great. I'm not expert with crossovers and is really helpful.
Cheers from Italy.
Maurizio
Hi Dave,
I should have prefaced the above with a major 'thank you', so apologies for not saying as much sooner. WinPCD is a terrific piece of software that I and many others use on a regular / constant basis, and all the work you've put in is very much appreciated. So in my best understated British mode: cheers for that, and if ever you find yourself in this neck of the woods, beers or your preferred choice are on me.
Hope the distortion rig etc. goes well -nothing like so involved, but I need to get my test baffle rebuilt over the next few weeks -when I went out to the garage last week, I found it had cracked top to bottom in the recent hot weather, which was a bit of a pain.
Scott
I should have prefaced the above with a major 'thank you', so apologies for not saying as much sooner. WinPCD is a terrific piece of software that I and many others use on a regular / constant basis, and all the work you've put in is very much appreciated. So in my best understated British mode: cheers for that, and if ever you find yourself in this neck of the woods, beers or your preferred choice are on me.
Understood -to be honest, I didn't expect it was particularly feasible (shows how much I know about programming!) so great to know it's at least possible for some point in the future. 🙂 Agreed, it probably wouldn't be a regular option for many, but definitely useful to have, & opens up a range of alternate circuits for more experienced designers.
Hope the distortion rig etc. goes well -nothing like so involved, but I need to get my test baffle rebuilt over the next few weeks -when I went out to the garage last week, I found it had cracked top to bottom in the recent hot weather, which was a bit of a pain.
Sounds interesting. I suppose I shouldn't ask what you have in mind for that, so I'll just repeat the thanks above!
Yeah, I was pondering that earlier & it sounds like a similar conclusion -perhaps some kind of 'advanced' button to enable / disable the extra element would work, just to avoid causing any confusion for less experienced users.
Scott
Glad to hear that you found the solution. The files look good to me. The impedance phase is correct based on the magnitude response. I checked that using SoundEasy (Bodzio Software), a commercial package, but well worth the price if you plan to do a lot more.
I put the spl file through my WinPhase app. The phase closely matches except at the lower end. The difference isn't significant and is based on extended the spl to 1 Hz for the calculation. The top end matches almost exactly. It can be done in SoundEasy, too, of course.
The primary reason I have used SoundEasy at times over the years is due to its digital filter capability. I you have a multi-channel amp, you can run a passive design to audition the design before constructing the actual crossover. Saves a huge amount of time when finalizing a design. And it includes the Ultimate Equalizer. I use for my main DSP dipole system. Check it out.
Dave