Hi to All,
I had some Baltic Birch ply left over from a previous project and decided to use it to build 40 liter stand mount 2 ways. The woofer section is set using Peerless 830869 8" woofers. I'm undecided on what route to go for the high frequencies.
My number one goal is to try and use parts I already have on hand yet achieve the best sound quality possible. In my limited experience, I have begun to understand that in speaker building there is certain compromises. With this design, it is either go the 8" + 1" mounted in a waveguide and push the Peerless woofer into an uncomfortable higher crossover frequency or go the route using the 3" full range and sacrifice the upper ranges somewhat.
Here is what I have for drivers:
Fountek Fr89ex 3" full range (2) already in enclosures (this would be easiest route to go)
Scanspeak D2604/833000 (1) would need to purchase 1 more plus necessary waveguides.
What choice would offer the best sound quality/ least compromises for a smallish room played at low to moderate levels.
Best Regards,
Rich
I had some Baltic Birch ply left over from a previous project and decided to use it to build 40 liter stand mount 2 ways. The woofer section is set using Peerless 830869 8" woofers. I'm undecided on what route to go for the high frequencies.
My number one goal is to try and use parts I already have on hand yet achieve the best sound quality possible. In my limited experience, I have begun to understand that in speaker building there is certain compromises. With this design, it is either go the 8" + 1" mounted in a waveguide and push the Peerless woofer into an uncomfortable higher crossover frequency or go the route using the 3" full range and sacrifice the upper ranges somewhat.
Here is what I have for drivers:
Fountek Fr89ex 3" full range (2) already in enclosures (this would be easiest route to go)
Scanspeak D2604/833000 (1) would need to purchase 1 more plus necessary waveguides.
What choice would offer the best sound quality/ least compromises for a smallish room played at low to moderate levels.
Best Regards,
Rich
Hi again, Rich. That's a pretty good 830869 woofer. I'd know what to do with that. 
Medium inductance. Pretty well-behaved. Reflex, around 30-40 L. I like to cross at about 3kHz, but increasingly like to notch or rolloff the 3kHz breakup nasties at 6kHz. It's the second harmonic that sounds rough, IMO
Not sure I greatly fancy a 86db 3" metal cone for some sort of top end, but looks like 2kHz crossover. And waveguide 1" soft-dome sounds like a lot of work. I would get interested in an 8" with an SB acoustics ring radiator. That could be good.
Medium inductance. Pretty well-behaved. Reflex, around 30-40 L. I like to cross at about 3kHz, but increasingly like to notch or rolloff the 3kHz breakup nasties at 6kHz. It's the second harmonic that sounds rough, IMO
Not sure I greatly fancy a 86db 3" metal cone for some sort of top end, but looks like 2kHz crossover. And waveguide 1" soft-dome sounds like a lot of work. I would get interested in an 8" with an SB acoustics ring radiator. That could be good.
Attachments
Last edited:
I am more inclined to a waveguided 1" HF unit crossed around 2 kHz but I would suggest first to explore the 3" wideband unit solution. My preference should not stand in your way anyway. These wide bands have their appeal and can be crossed lower where human hearing tolerates more of the crossover design mistakes without being terribly wrong. Have fun!
Thanks so much for the quick replies! Great to hear from you both. I really value the expertise that you both have in speaker design. If I was inclined to go the waveguide route, would the scanspeak d2604 tolerate a crossover at 2 khz with 3rd or 4th order electrical on tweeter? It seems you both believe the woofer is well behaved enough to cross higher then I originally thought, so might gut tells me I should take on the waveguide route. I know it will have its challenges, but also loads of fun!
Best,
Rich
Best,
Rich
According to Zaph the Vifa single magnet counterpart can be crossed as low as 1.8KHz LR4 acoustic. A proprer waveguide can only do things better. The Vifa and SS tweeters should have the same faceplate so have a look here, there is also a link to a measurement: Waveguide for a Vifa tweeter
Ralf
Ralf
TBH, these days, I would do a serious crossover on 8" to 1" or whatever.
Let's look at Alan Shaw's work at Harbeth with the Compact 7 ES-3.
The reason Harbeth can charge the big bucks for their designs, is the sheer quality of design. This speaker will play the Berlin Philarmonic on a grand scale with low distortion. Polycone and metal dome here. But high order crossover.
And, doubtless some serious impedance equalisation. I just acknowledge BBC greatness sometimes. This is superb. And weirdly, I have never heard them.
Let's look at Alan Shaw's work at Harbeth with the Compact 7 ES-3.
The reason Harbeth can charge the big bucks for their designs, is the sheer quality of design. This speaker will play the Berlin Philarmonic on a grand scale with low distortion. Polycone and metal dome here. But high order crossover.
And, doubtless some serious impedance equalisation. I just acknowledge BBC greatness sometimes. This is superb. And weirdly, I have never heard them.
Attachments
CONSIDER: 1200Hz waveguide OR robust dome with steep LR4-LR4 circuits at 1400Hz.
==========
The Peerless 830869 looks like a good woofer to mate with a waveguide dome tweeter designed for a 1200Hz crossover. Visit PartsExpress and diysoundgroup to check their dome-based waveguides and "robust" dome tweeters. Controlled Directivity.
The diysoundgroup $17 SEOS-8 waveguide was designed to use with the $17 Dayton Audio RS28 with a 1200Hz crossover.
==========
The SB-Acoustics SB29RDC 1.1" dome tweeter is robust enough to support an LR4 crossover circuit at 1400Hz, where the Peerless 830869 datasheet shows very modest early beaming mainly at 60-degrees.
===========
Despite an Fs=475Hz, the SPL data for the ScanSpeak 833000 tweter suggests a crossover frequency closer to 2kHz, which is not a good match for the Peerless 830869.
==========
The Peerless 830869 looks like a good woofer to mate with a waveguide dome tweeter designed for a 1200Hz crossover. Visit PartsExpress and diysoundgroup to check their dome-based waveguides and "robust" dome tweeters. Controlled Directivity.
The diysoundgroup $17 SEOS-8 waveguide was designed to use with the $17 Dayton Audio RS28 with a 1200Hz crossover.
==========
The SB-Acoustics SB29RDC 1.1" dome tweeter is robust enough to support an LR4 crossover circuit at 1400Hz, where the Peerless 830869 datasheet shows very modest early beaming mainly at 60-degrees.
===========
Despite an Fs=475Hz, the SPL data for the ScanSpeak 833000 tweter suggests a crossover frequency closer to 2kHz, which is not a good match for the Peerless 830869.
Attachments
Greets!
Well, just from published specs and using cone/domes in a KISS system, I would want a ~2.86" dia. piston, so already having the 3" makes it a 'no brainer' for me to cobble together with a 500-1200 Hz XO depending on what sounds best overall.
GM
GM,
Thanks for your input! I may indeed go that route it would certainly save a lot of time and effort.
Best,
Rich
You're welcome! Keep them in separate cabs to physically time align them, so that you won't have to reversing the HF's polarity on a 2nd order XO.
GM
Hey Rich,
I think if you were running these at high volumes, the 8" + 3" would be the best choice. The FR unit won't have any trouble keeping up with a xo probably as low as 400 or 500Hz (though I might go just a little higher than that) and you'll stay the farthest away from the woofer's breakup. On the con side as with any FR unit, off-axis response suffers more in the HF's. You have them in boxes right now - the midrange is supposed to be very good on those so I guess the question is are you fully satisfied with symbols and so forth on the top end?
On the other hand, if you don't push them too hard, it doesn't look like there is any reason why the 8" and that tweeter aren't going to work nicely. From Zaph re the D2604: "You could probably cross over as low as you wanted, though if you go down to 1.5kHz I wouldn't count on it getting too loud. Maybe 2.2 or 2.4 kHz LR2 could work." And that's without a waveguide. For the woofer, off-axis response looks fine at up to about 2,000Hz. So I guess the big question is how bad is that woofer breakup around 3000Hz?
I would look at a sim with LR4 acoustic at between 1600 -1800Hz with a LCR resonance compensation filter on the tweeter and with a steep tanking or notch filter on the woofer's peak resonance. LineSource's sim is down about 18dB at 3000Hz but just to be on the safe side, I would try to push it down even further. And going with a waveguide will only make things better.
And then if you have some spare parts sitting around, I would consider perhaps putting 1 of each together and seeing which one you like best.
I think if you were running these at high volumes, the 8" + 3" would be the best choice. The FR unit won't have any trouble keeping up with a xo probably as low as 400 or 500Hz (though I might go just a little higher than that) and you'll stay the farthest away from the woofer's breakup. On the con side as with any FR unit, off-axis response suffers more in the HF's. You have them in boxes right now - the midrange is supposed to be very good on those so I guess the question is are you fully satisfied with symbols and so forth on the top end?
On the other hand, if you don't push them too hard, it doesn't look like there is any reason why the 8" and that tweeter aren't going to work nicely. From Zaph re the D2604: "You could probably cross over as low as you wanted, though if you go down to 1.5kHz I wouldn't count on it getting too loud. Maybe 2.2 or 2.4 kHz LR2 could work." And that's without a waveguide. For the woofer, off-axis response looks fine at up to about 2,000Hz. So I guess the big question is how bad is that woofer breakup around 3000Hz?
I would look at a sim with LR4 acoustic at between 1600 -1800Hz with a LCR resonance compensation filter on the tweeter and with a steep tanking or notch filter on the woofer's peak resonance. LineSource's sim is down about 18dB at 3000Hz but just to be on the safe side, I would try to push it down even further. And going with a waveguide will only make things better.
And then if you have some spare parts sitting around, I would consider perhaps putting 1 of each together and seeing which one you like best.
Hi jReave,
Great to hear from you! These speakers will not be played a loud listening levels. If such is required, I could plug in the 3 ways you helped me design earlier this year. Those will play loud enough to change my neighbor's taste in music. And my nearest neighbor is a half mile away.
My only dilemma in trying the A/B test you suggested is if I tried the tweeter/ waveguide solution it would require router work on the woofer cab as I intended to go with both woofer and tweeter sharing the same box if going that route. This would give me the desired center to center spacing between drivers. Otherwise, using the fullrangers would only require placing those on top of woofer cabs.
I'm leaning toward going the tweeter/ waveguide combo. This would also give me the chance to try something new. I will study on the sim that LineSource posted for help in the crossover design.
Thanks so much for your outstanding comments!
Best Regards,
Rich
Great to hear from you! These speakers will not be played a loud listening levels. If such is required, I could plug in the 3 ways you helped me design earlier this year.
Those will play loud enough to change my neighbor's taste in music. And my nearest neighbor is a half mile away.My only dilemma in trying the A/B test you suggested is if I tried the tweeter/ waveguide solution it would require router work on the woofer cab as I intended to go with both woofer and tweeter sharing the same box if going that route. This would give me the desired center to center spacing between drivers. Otherwise, using the fullrangers would only require placing those on top of woofer cabs.
I'm leaning toward going the tweeter/ waveguide combo. This would also give me the chance to try something new. I will study on the sim that LineSource posted for help in the crossover design.
Thanks so much for your outstanding comments!
Best Regards,
Rich
For testing, I guess I was thinking about a detachable baffle. And scrap wood. I may not even bother with the waveguide for testing purposes. So yes, route that out and give it a shot. And then disconnect the tweeter and place the FR on top. See how that goes. (Well and change the xo)
Depending on the choices you make, you're only out a small piece of wood and some time, I would think.
Depending on the choices you make, you're only out a small piece of wood and some time, I would think.
I vote for the 3 inch mid-tweeter. The only real downside of that is that the treble rolls off when you're off axis. I actually like that since a lot of the music I listen to has pumped that up a bit. There's also the issue that the 3 inch drivers I love (Peerless TG9FD1008) rolls off at about 15kHZ. I don't miss the energy above that at all. Maybe because I'm 62 and don't hear perfectly anymore.
On the good side, the off axis frequency response will have a much more gradual transition at the crossover frequency because the size of the driver diaphrams won't change so drastically. So the way the room will react to the acoustic energy will be better. Also, the cone breakup resonance that most woofers have is something you want to bury deeply. So being able to have a crossover at say 500HZ instead of 3kHZ allows you to push the 4-6kHZ (whatever) woofer resonance way down.
Lynn Olson once mentioned that dome tweeters can have coil tilt I.M. distortion problems when you cross them over too low. And that is at frequencies where the ear is most sensitive (1kHZ - 5kHZ - See Fletcher-Munson graphs).
I built a bookshelf speaker system that has a 8 inch kevlar cone woofer in a sealed cabinet (with active EQ making it acoustically flat to 30HZ) (active meaning a circuit ahead of the poweramp), and the 3 inch Peerless TG9FD1008 driver in a sub-enclosure handling from 500HZ on up. I used a 4 pole active crossover, so it's biamp'd. I think it sounds fantastic, and it's one of the simplest speaker systems I've ever made. Here's a link to it on my hobby website if you're interested: Untitled Document
On the good side, the off axis frequency response will have a much more gradual transition at the crossover frequency because the size of the driver diaphrams won't change so drastically. So the way the room will react to the acoustic energy will be better. Also, the cone breakup resonance that most woofers have is something you want to bury deeply. So being able to have a crossover at say 500HZ instead of 3kHZ allows you to push the 4-6kHZ (whatever) woofer resonance way down.
Lynn Olson once mentioned that dome tweeters can have coil tilt I.M. distortion problems when you cross them over too low. And that is at frequencies where the ear is most sensitive (1kHZ - 5kHZ - See Fletcher-Munson graphs).
I built a bookshelf speaker system that has a 8 inch kevlar cone woofer in a sealed cabinet (with active EQ making it acoustically flat to 30HZ) (active meaning a circuit ahead of the poweramp), and the 3 inch Peerless TG9FD1008 driver in a sub-enclosure handling from 500HZ on up. I used a 4 pole active crossover, so it's biamp'd. I think it sounds fantastic, and it's one of the simplest speaker systems I've ever made. Here's a link to it on my hobby website if you're interested: Untitled Document
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.