Heehee, I was off yesterday and today. I must say though- I needed to be. Got done loading around 9 this morning- before the rain started. I had tracks ripped yesterday, whittled down the timeframes late last night into this morning of each to what I selected about a week ago, made sure levels were set correctly by about 7:30, and burned 15 copies by 10am.
I am planning to be there in the tomorrow before lunch. Was hoping for 9ish, but more likely 10 or 11. Crossovers are finally done but I still need to finish mounting tweeters in the B19's, quite possibly happening in the AM. Can't wait to see everyone and hear some great sounds.
Just filled up for $3.94/gallon (SE Wisconsin). Bet I won't see that price again for a few days. Plan to get there 1 to 3pm or so tomorrow. Retro Speako's and xovers are packed and ready to go.
Hi Ben, Can you do a last minute addition for me. I would be starting today and will be there for Saturday only.
I am going to bring an active bookshelf speaker (non contest). Would it be possible to get the house track at the venue on usb or somehow copy to my laptop? The active speakers run through my laptop.
Fantastic time! Thanks for all the kind comments on my speakers. Such a great batch of projects showed up and was awesome to see all you guys/gals again. Hope everyone makes it home safe! Y'all rock!!!🤘🏼
Always great to see people frome the forum. Eric, you might not remember but the first time we met we were laughing about "hifi" USB cables and your comment on how they snuck a 2 between all the ones and zeros. Still one of the best hifi jokes I've ever heard. It was great to see you again, and of course the rest of the crew.
Ha! Nope, I don't remember that, but it's definitely a smart ass remark I'd make😆
Man, it was such a great time hanging with everyone and meeting new geeks! I just got in and Adam is on his way home. Thanks to who ever supplied the swag, I'll definitely start a build with those MCM 8's. I can say enough how awesome it is to have friends like you all and can't wait to do it again.🤘🏼
Looks like a bunch of fun speakers! Now that the event/challenge is over, I'm curious to know if anyone else noticed the ~20mm of phase error in the woofer polar data provided? I'll share my frustrations with it if anyone is interested.
I knew about that. You're talking at the low-end, yes? It's an artifact of the ARTA (full package) measurements, and since it doesn't affect the usage in Vituixcad since it will remove it, Coop didn't see the issue. He has contacted the guy behind ARTA, and the guy seems to not care to fix the issue.
Jack and I were the only 2 that voiced the xover. All of the others were from the data. FWIW, dcibel went first after the 'warts n all' off the shelf network I hooked up for a baseline. We had 11 entries. I was number 7, EQ from Singapore was number 2, and Jack was number 9- those 3 ended up in the final round.
EQ won the competition. He designed his to stand out with good power response, an an inherent BBC broad dip. Jack didn't comp much, but had flat response with a boosted BSC and extended bass. I fixed all the warts with really flat response utilizing 19 parts.
Most of the entries howled under the opera track, and I thought EQ and mine were the only 2 that did not.
It was tough, but insanely fun, and noone was the wiser to which was their own design or their friend's design.
No, I'm not talking about the low end, I'm talking about the entire set of responses is off by about 20mm when compared to the "for offset" measurement at the tweeter axis. Based on these findings I would expect there to be some real differences in design for those who used the "for offset" measurement data with software like Xsim, vs those who used the polar data set.
Of course, the obvious difference is that the "for offset" measurement is taken on the tweeter axis, and the polar data is on the driver axis, so there should be some difference between the two. There is 145mm between drivers and 1m measurement distance, so the woofer is 8.25 degrees off-axis for the "for offset" measurement, and the distance from mic to speaker baffle is 1010.5mm for the woofer, or 10.5mm further than it is to the tweeter. The delay determination using the "for offset" measurements shows no delay required, so a dual channel (or at least semi-dual) measurement system is assumed to be used with locked FFT window start time.
With the above information in mind, I compared the vertical 10 degree polar measurement, as it is closer to the 8.25 angle for direct comparison. The frequency response is nearly exactly the same, unfortunately the phase difference between the two files is 90us, or 30mm, not the around 10mm difference that is expected. This is easily observed using VituixCAD, I used the calculator tool. Simply load both response files in the "A responses" section. Save the phase response of one as an overlay, then adjust delay of the other until you achieve an overlap.
The assumptions that can be made is that either the mic has moved, the speaker cabinet has moved, for the FFT window start location has moved, or poor use of a measuring tape. Due to this error in phase, in order to achieve the same result using the polar data and the tweeter axis "for offset" measurements, you would require a 20mm delay to be added to the woofer measurement set.
Going back to my original comparison of the "for offset" vs the polar data, I hope @kenrhodes doesn't mind if I use his design as an example since its a bit more dramatic than my own design. Ken used Xsim, and correctly used the "for offset" response files with no delay included. The intended response in this configuration is below:
Now, the expectation would be a nearly identical response by using the polar vertical 10 degree measurement, as well as inclusion of a 10mm delay as determined above. However the result is not equal.
For confirmation of the phase error, yes we do get a near equal result if 30mm of delay is included instead.
For my design, the difference in result was somewhat minimal, and I had already sent it off to Meniscus for them to build, so I decided to leave it as-is and see what comes of it. However just prior to the event Ken Rhodes shared with me his design, and the difference in response in his crossover is actually quite significant depending on the measurement data used, as you can see above.
Comments
Thinking we'll be there afternoon
Freaking work is going to slow today😛
Same, we're targeting 3:00 PM
Heehee, I was off yesterday and today. I must say though- I needed to be. Got done loading around 9 this morning- before the rain started. I had tracks ripped yesterday, whittled down the timeframes late last night into this morning of each to what I selected about a week ago, made sure levels were set correctly by about 7:30, and burned 15 copies by 10am.
I've been busy for sure....
InDIYana Event Website
I am planning to be there in the tomorrow before lunch. Was hoping for 9ish, but more likely 10 or 11. Crossovers are finally done but I still need to finish mounting tweeters in the B19's, quite possibly happening in the AM. Can't wait to see everyone and hear some great sounds.
See you guys in roughly 20 hours.
Check your PM's, Ben.
Just filled up for $3.94/gallon (SE Wisconsin). Bet I won't see that price again for a few days. Plan to get there 1 to 3pm or so tomorrow. Retro Speako's and xovers are packed and ready to go.
I have no idea what a gallon of go juice costs on a plane, but I'm sure we burned plenty getting to Ft. Wayne.
Did they serve you an inflight snack and beverage?
They tried, but I declined. Three peanuts and a shot of Sprite just doesn't appeal to me.
The hop from Chicago to Ft. Wayne is all of 30 minutes - never even hit cruising altitude.
Have a great time everyone. Please send lots of pics. Y'all got my number.
Hi Ben, Can you do a last minute addition for me. I would be starting today and will be there for Saturday only.
I am going to bring an active bookshelf speaker (non contest). Would it be possible to get the house track at the venue on usb or somehow copy to my laptop? The active speakers run through my laptop.
Ani, Ben has cd's available will that work?
no cd player in my laptop
I have other tracks i can play.
I am trying the hifiberry's Beocreate All in one, 4 channel DSP amp, the input options are limited.
We will work on getting you a digital option. Looking forward to seeing you again. Safe travels.
Fantastic time! Thanks for all the kind comments on my speakers. Such a great batch of projects showed up and was awesome to see all you guys/gals again. Hope everyone makes it home safe! Y'all rock!!!🤘🏼
Always great to see people frome the forum. Eric, you might not remember but the first time we met we were laughing about "hifi" USB cables and your comment on how they snuck a 2 between all the ones and zeros. Still one of the best hifi jokes I've ever heard. It was great to see you again, and of course the rest of the crew.
Ha! Nope, I don't remember that, but it's definitely a smart ass remark I'd make😆
Man, it was such a great time hanging with everyone and meeting new geeks! I just got in and Adam is on his way home. Thanks to who ever supplied the swag, I'll definitely start a build with those MCM 8's. I can say enough how awesome it is to have friends like you all and can't wait to do it again.🤘🏼
...oh and the A-Team rules!!!😛
InDIYana 2022
Definitely a good time. Eric, your speakers are one of the coolest looking designs ever, and damn fine sounding as well. Yes, the A-team rules!!!!
Link to the demo I put together for the theme event:
https://1drv.ms/u/s!AlHuPRQmBiX2hu1q0pKAJAQIcMzTAg?e=4eTBr3
I recommend listening to it on some nice cans, or with a system that can resolve sub-30Hz bass.
Thank you again for doing that. They suited the theme really well.
InDIYana Event Website
Yep, awesome tracks!
I just played the last track on my big "daily driver" speaker that have an f3 of around 20hz and it pressurized the room, badass!
Looks like a bunch of fun speakers! Now that the event/challenge is over, I'm curious to know if anyone else noticed the ~20mm of phase error in the woofer polar data provided? I'll share my frustrations with it if anyone is interested.
I knew about that. You're talking at the low-end, yes? It's an artifact of the ARTA (full package) measurements, and since it doesn't affect the usage in Vituixcad since it will remove it, Coop didn't see the issue. He has contacted the guy behind ARTA, and the guy seems to not care to fix the issue.
InDIYana Event Website
...and about the competition....
Jack and I were the only 2 that voiced the xover. All of the others were from the data. FWIW, dcibel went first after the 'warts n all' off the shelf network I hooked up for a baseline. We had 11 entries. I was number 7, EQ from Singapore was number 2, and Jack was number 9- those 3 ended up in the final round.
EQ won the competition. He designed his to stand out with good power response, an an inherent BBC broad dip. Jack didn't comp much, but had flat response with a boosted BSC and extended bass. I fixed all the warts with really flat response utilizing 19 parts.
Most of the entries howled under the opera track, and I thought EQ and mine were the only 2 that did not.
It was tough, but insanely fun, and noone was the wiser to which was their own design or their friend's design.
Good times!
InDIYana Event Website
No, I'm not talking about the low end, I'm talking about the entire set of responses is off by about 20mm when compared to the "for offset" measurement at the tweeter axis. Based on these findings I would expect there to be some real differences in design for those who used the "for offset" measurement data with software like Xsim, vs those who used the polar data set.
Of course, the obvious difference is that the "for offset" measurement is taken on the tweeter axis, and the polar data is on the driver axis, so there should be some difference between the two. There is 145mm between drivers and 1m measurement distance, so the woofer is 8.25 degrees off-axis for the "for offset" measurement, and the distance from mic to speaker baffle is 1010.5mm for the woofer, or 10.5mm further than it is to the tweeter. The delay determination using the "for offset" measurements shows no delay required, so a dual channel (or at least semi-dual) measurement system is assumed to be used with locked FFT window start time.
With the above information in mind, I compared the vertical 10 degree polar measurement, as it is closer to the 8.25 angle for direct comparison. The frequency response is nearly exactly the same, unfortunately the phase difference between the two files is 90us, or 30mm, not the around 10mm difference that is expected. This is easily observed using VituixCAD, I used the calculator tool. Simply load both response files in the "A responses" section. Save the phase response of one as an overlay, then adjust delay of the other until you achieve an overlap.
The assumptions that can be made is that either the mic has moved, the speaker cabinet has moved, for the FFT window start location has moved, or poor use of a measuring tape. Due to this error in phase, in order to achieve the same result using the polar data and the tweeter axis "for offset" measurements, you would require a 20mm delay to be added to the woofer measurement set.
Going back to my original comparison of the "for offset" vs the polar data, I hope @kenrhodes doesn't mind if I use his design as an example since its a bit more dramatic than my own design. Ken used Xsim, and correctly used the "for offset" response files with no delay included. The intended response in this configuration is below:
Now, the expectation would be a nearly identical response by using the polar vertical 10 degree measurement, as well as inclusion of a 10mm delay as determined above. However the result is not equal.
For confirmation of the phase error, yes we do get a near equal result if 30mm of delay is included instead.
For my design, the difference in result was somewhat minimal, and I had already sent it off to Meniscus for them to build, so I decided to leave it as-is and see what comes of it. However just prior to the event Ken Rhodes shared with me his design, and the difference in response in his crossover is actually quite significant depending on the measurement data used, as you can see above.