Hi all.
I’ve had another thought to the whole low crossover scenario. Instead of the wavecor waveguide tweeter or sb29 tweeter, what about using a full range? I’m thinking of the visaton b80, and crossover at 500hz. I can sacrifice 1l for an enclosure and then use a 19mm tweeter if needed.
Thoughts everyone?
I’ve had another thought to the whole low crossover scenario. Instead of the wavecor waveguide tweeter or sb29 tweeter, what about using a full range? I’m thinking of the visaton b80, and crossover at 500hz. I can sacrifice 1l for an enclosure and then use a 19mm tweeter if needed.
Thoughts everyone?
Hi All.
I have some baffle data now and have mocked up a crossover using the SB29 tweeter as a 2 way. To my ears it just doesn't sound quite right.
Attached are some screen shots of a few scenarios with the B80 as a wide band, B80 and SB29, and just the SB29. The SB23NBACS driver requires way too many components to pull it into line with the 1.1 - 1.2K crossover.
As I haven't purchased the mid drivers yet, I can only go on published data. I even had a go at Boxsim (a fav of system7) today and it's not too bad to use, although get conflicting results using the same frd and zma files.
All crossovers look OK, but what do they sound like?????
My boxes are almost complete and now contemplating what to do with the baffles before I start with the router.
Comments appreciated.
I have some baffle data now and have mocked up a crossover using the SB29 tweeter as a 2 way. To my ears it just doesn't sound quite right.
Attached are some screen shots of a few scenarios with the B80 as a wide band, B80 and SB29, and just the SB29. The SB23NBACS driver requires way too many components to pull it into line with the 1.1 - 1.2K crossover.
As I haven't purchased the mid drivers yet, I can only go on published data. I even had a go at Boxsim (a fav of system7) today and it's not too bad to use, although get conflicting results using the same frd and zma files.
All crossovers look OK, but what do they sound like?????
My boxes are almost complete and now contemplating what to do with the baffles before I start with the router.
Comments appreciated.
Attachments
Hi all.
So I’ve bitten the bullet and purchased the wavecor TW030WA12 drivers. They’ll give me a bit more headroom to cross at 1khz if required. I’ve managed to get the sb23nbacs driver sounding good. 4th order electrical crossover with a notch filter.
I have a couple of questions with regard to my test data. I have followed the guide on audio judgement on how to find the acousic centre and have put that data into Xsim. I have created a crossover and then measured with REW. Xsim puts the tweeter in electrical phase with the woofer to get a flat response. When I measure in REW with drivers in electrical phase there’s a huge suckout at 1.2khz. When i reverse the tweeter connections and measure again, I get a flat response. Is this normal?
So I’ve bitten the bullet and purchased the wavecor TW030WA12 drivers. They’ll give me a bit more headroom to cross at 1khz if required. I’ve managed to get the sb23nbacs driver sounding good. 4th order electrical crossover with a notch filter.
I have a couple of questions with regard to my test data. I have followed the guide on audio judgement on how to find the acousic centre and have put that data into Xsim. I have created a crossover and then measured with REW. Xsim puts the tweeter in electrical phase with the woofer to get a flat response. When I measure in REW with drivers in electrical phase there’s a huge suckout at 1.2khz. When i reverse the tweeter connections and measure again, I get a flat response. Is this normal?
Hi,
Well done.
The Wavecor TW030WA12 certainly is the best option for a low crossed 2-way. However, I wouldn't go below 1.5 kHz nevertheless.
The phase issue isn't normal since wave length at 1.2 kHz is 28 cm and an error of 14 cm is unlikely. Without knowing your baffle layout and your crossover I can't say more.
Are you working with actual measurements now? I noticed that HiFiCompass only lists the TW030WA11, that's the 4 ohm version of the tweeter, but not the TW030WA12.
Well done.
The Wavecor TW030WA12 certainly is the best option for a low crossed 2-way. However, I wouldn't go below 1.5 kHz nevertheless.The phase issue isn't normal since wave length at 1.2 kHz is 28 cm and an error of 14 cm is unlikely. Without knowing your baffle layout and your crossover I can't say more.
Are you working with actual measurements now? I noticed that HiFiCompass only lists the TW030WA11, that's the 4 ohm version of the tweeter, but not the TW030WA12.
Thanks Dissi,
My baffle is 90x30cm. Woofer located 60cm from bottom and centred. tweeter is located 77cm from bottom and offset by 5cm from centre. The data I have at present is actual data with the SB drivers. Separate measurements taken on axis with the tweeter at 1m from the drivers and 1.2m from floor. Each driver measured and then with the drivers in parallel. Used REW, 200Hz-20KHz with 5ms window and then 1/24 octave smoothing. Put that data into Xsim and measured the acoustic offset. The crossover is an iteration of one that is posted here. 3rd order electrical for woofer, and 4th order electrical for tweeter. Crossover point is approx 1.2KHz. Acoustic rolloff is 6-8th order.
Now does that mean when I measure on the flat baffle, and include the offset, I will have the phase issue and have to reverse the tweeter? If I tilt the baffle back, say 10 degrees, so the offset is nullified, will I be able to have the drivers in electrical phase?
If I don't put the acoustic offset into the crossover, the tweeter has to be wired in reverse polarity to produce a flat response. I wanted to do this to learn before the Wavecor drivers turn up.
My next question is whether to offset the Wavecor tweeter or just leave it in the centre directly above the woofer and chamfer the edges. If I offset, I will just use a roundover bit on the baffle.
Cheers
My baffle is 90x30cm. Woofer located 60cm from bottom and centred. tweeter is located 77cm from bottom and offset by 5cm from centre. The data I have at present is actual data with the SB drivers. Separate measurements taken on axis with the tweeter at 1m from the drivers and 1.2m from floor. Each driver measured and then with the drivers in parallel. Used REW, 200Hz-20KHz with 5ms window and then 1/24 octave smoothing. Put that data into Xsim and measured the acoustic offset. The crossover is an iteration of one that is posted here. 3rd order electrical for woofer, and 4th order electrical for tweeter. Crossover point is approx 1.2KHz. Acoustic rolloff is 6-8th order.
Now does that mean when I measure on the flat baffle, and include the offset, I will have the phase issue and have to reverse the tweeter? If I tilt the baffle back, say 10 degrees, so the offset is nullified, will I be able to have the drivers in electrical phase?
If I don't put the acoustic offset into the crossover, the tweeter has to be wired in reverse polarity to produce a flat response. I wanted to do this to learn before the Wavecor drivers turn up.
My next question is whether to offset the Wavecor tweeter or just leave it in the centre directly above the woofer and chamfer the edges. If I offset, I will just use a roundover bit on the baffle.
Cheers
Hi all.
I have the wavecor tweeters in my hot hands. They are huge compared to the sb29rdcn tweeters. I’ve put the two through dats software and they came up very similar. Fs of 450hz, slightly higher than manufacture data, but all other parameters very close.
The baffles are almost complete, and I have assembled one of the boxes. Testing won’t be far off. Can’t wait for actual data with the drivers in the box.
Cheers.
I have the wavecor tweeters in my hot hands. They are huge compared to the sb29rdcn tweeters. I’ve put the two through dats software and they came up very similar. Fs of 450hz, slightly higher than manufacture data, but all other parameters very close.
The baffles are almost complete, and I have assembled one of the boxes. Testing won’t be far off. Can’t wait for actual data with the drivers in the box.
Cheers.
Hi all.
I've managed to get one of the boxes to a semi completed stage and get some actual data. I have a question about baffle faceting and diffraction that I hope can be answered by someone who knows more about it than me. There seems to be some diffraction at 2 and 4khz, could this be due to the top chamfer?
I'm aware of the wavy response down low. It's due to the unfinished T line.
Attached is a pic of the baffle and a mock up crossover using Boxsim.
Cheers
I've managed to get one of the boxes to a semi completed stage and get some actual data. I have a question about baffle faceting and diffraction that I hope can be answered by someone who knows more about it than me. There seems to be some diffraction at 2 and 4khz, could this be due to the top chamfer?
I'm aware of the wavy response down low. It's due to the unfinished T line.
Attached is a pic of the baffle and a mock up crossover using Boxsim.
Cheers
Attachments
Hi all.
So it has been a rainy old day today so I thought I'd get some more data in between showers. I took some more measurements after faceting the top chamfer. It's now 40x16, instead of a 16mm chamfer. Took measurements of the mid bass and terminus nearfield and merged with the gated response. I must say that the bass that comes out of this enclosure is quite astonishing.
I'll update more soon.
Cheers
So it has been a rainy old day today so I thought I'd get some more data in between showers. I took some more measurements after faceting the top chamfer. It's now 40x16, instead of a 16mm chamfer. Took measurements of the mid bass and terminus nearfield and merged with the gated response. I must say that the bass that comes out of this enclosure is quite astonishing.
I'll update more soon.
Cheers
Hi all.
I've been doing some crossover work and have run into a couple of issues with some of the simulator software packages.
What I did was put the same data into 4 different packages. Win PCD, Xsim, Boxsim, and Virtuix CAD. I created a crossover with 2nd order electrical for woofer and 3rd order electrical for tweeter. Ran the optimiser in Virtuix CAD and Boxsim and they both came up with different results.
Attached are some screen shots of the programs.
I also created a mockup crossover and took some measurements in REW. Also attached with on axis and 30 deg off axis response.
Cheers
I've been doing some crossover work and have run into a couple of issues with some of the simulator software packages.
What I did was put the same data into 4 different packages. Win PCD, Xsim, Boxsim, and Virtuix CAD. I created a crossover with 2nd order electrical for woofer and 3rd order electrical for tweeter. Ran the optimiser in Virtuix CAD and Boxsim and they both came up with different results.
Attached are some screen shots of the programs.
I also created a mockup crossover and took some measurements in REW. Also attached with on axis and 30 deg off axis response.
Cheers
Attachments
Progress update
Hi all.
Just an update on the progress of the boxes. Both are now completed and have 2 coats of primer sealer applied. I managed to make up a jig and chamfer the front panels better than I did first time around. The tweeter data looks better now. Attached is a pic of the boxes and current crossover simulation with actual measured data. The bass hump is from merged data from the terminus. It's not really that pronounced.
Hi all.
Just an update on the progress of the boxes. Both are now completed and have 2 coats of primer sealer applied. I managed to make up a jig and chamfer the front panels better than I did first time around. The tweeter data looks better now. Attached is a pic of the boxes and current crossover simulation with actual measured data. The bass hump is from merged data from the terminus. It's not really that pronounced.
Attachments
Progress update
Hi all,
The boxes are now complete and I have some new measurement data. The latest version of the crossover has moved the crossover frequency to 1650 Hz. Up from 1350 Hz with previous versions.
I've scrounged up enough components to create a couple of real ones to compare and I must say that the latest one sounds way better. No lobing, off axis response is excellent, and the power response is decent as well.
The lower crossover point of the original made the tweeter sound strained.
Now I have to ask a serious question about crossover caps. What is everyone's preference?
I have chosen a mixture of caps for these speakers. Mundorf Mcap EVO oil for the tweeter, Jantzen standard Z cap and Mundorf Mcap EVO for the woofer. Am I on the right track?
I'm thinking of adding Cornell Dubilier 940C bypass caps to the tweeter circuit. I've read that these can be of benefit to squeeze a bit more detail out of the tweeter. Is this true?
Cheers
Hi all,
The boxes are now complete and I have some new measurement data. The latest version of the crossover has moved the crossover frequency to 1650 Hz. Up from 1350 Hz with previous versions.
I've scrounged up enough components to create a couple of real ones to compare and I must say that the latest one sounds way better. No lobing, off axis response is excellent, and the power response is decent as well.
The lower crossover point of the original made the tweeter sound strained.
Now I have to ask a serious question about crossover caps. What is everyone's preference?
I have chosen a mixture of caps for these speakers. Mundorf Mcap EVO oil for the tweeter, Jantzen standard Z cap and Mundorf Mcap EVO for the woofer. Am I on the right track?
I'm thinking of adding Cornell Dubilier 940C bypass caps to the tweeter circuit. I've read that these can be of benefit to squeeze a bit more detail out of the tweeter. Is this true?
Cheers
Attachments
Progress update
Hi All,
Sorry about yesterdays post. It contained an incorrect screen snapshot of some of my original data. It certainly doesn't look like that now.
I now have two versions of the crossover that I'd like some feedback on from those here with way more knowledge than me. I'm not sure if the 4K breakup has been tamed enough in the first one.
First version is a take on the Seas TRYM woofer circuit (thanks System 7), and the other is all my own. Both tweeter circuits are the same. Crossover point is 1630Hz with a sharp null when inverting tweeter polarity.
I believe that I'm getting real close to a finished product.
Cheers
Hi All,
Sorry about yesterdays post. It contained an incorrect screen snapshot of some of my original data. It certainly doesn't look like that now.
I now have two versions of the crossover that I'd like some feedback on from those here with way more knowledge than me. I'm not sure if the 4K breakup has been tamed enough in the first one.
First version is a take on the Seas TRYM woofer circuit (thanks System 7), and the other is all my own. Both tweeter circuits are the same. Crossover point is 1630Hz with a sharp null when inverting tweeter polarity.
I believe that I'm getting real close to a finished product.
Cheers
Attachments
Last edited:
Both designs seem to have very similar responses and shapes. As you say they both give sharp nulls which suggests how close they are. I think it would be difficult to hear the 4k output as it is 25 or so dBs down. I guess it is time to listen and tweak if needed. I imagine that you have the basis of something good with either one. An even simpler 1.8 and 16uF might not sound to bad.
I don't know how it looks in the simulation with virtuix.
I don't know how it looks in the simulation with virtuix.
Thanks Raymondj.
I think I have enough components to put together the crossover in the first attachment, and already have the 2nd one mocked up. I've been testing it with just the radio on my receiver and it definitely sounds like a single point source. Very well behaved. Deep bass, articulate vocals, mellow mids, and crisp but not bright highs.
I'll run a sim when I get home from work with just the 16uf cap on the woofer and see how it all correlates.
Pity I have already purchased some components, but they were mainly for the tweeter network, as it seems solid. The simpler the network the better in my books.
Cheers
I think I have enough components to put together the crossover in the first attachment, and already have the 2nd one mocked up. I've been testing it with just the radio on my receiver and it definitely sounds like a single point source. Very well behaved. Deep bass, articulate vocals, mellow mids, and crisp but not bright highs.
I'll run a sim when I get home from work with just the 16uf cap on the woofer and see how it all correlates.
Pity I have already purchased some components, but they were mainly for the tweeter network, as it seems solid. The simpler the network the better in my books.
Cheers
Hi eg92b16a,
It's good that you have some extra components. I can guarantee having come this far you will try your circuits and then decide which is best. In the end you will prefer one and it will only be possibly 0.5-1.5dB different from the others. Once you are happy with the bass, you might concentrate on listening to the treble and listen and measure that with just one value change. I haven't read the whole thread can you do responses for 30 and 45 degrees of axis as ideally these should be gently decaying in a smooth fashion.
It's good that you have some extra components. I can guarantee having come this far you will try your circuits and then decide which is best. In the end you will prefer one and it will only be possibly 0.5-1.5dB different from the others. Once you are happy with the bass, you might concentrate on listening to the treble and listen and measure that with just one value change. I haven't read the whole thread can you do responses for 30 and 45 degrees of axis as ideally these should be gently decaying in a smooth fashion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.