You can read the discussion of the Tysen speaker on this thread:
http://www.diyaudio.com/forums/showthread.php?t=144099
But I decided, instead of putting my project completely onto that thread, I should start my own.
Anyway, onto the project.
Seems appropriate that I start woth driver choice, so I will.
The FR driver is a Visaton FRS8 - these are (in terms of price to performance), fantastic speakers, which sacrifice (seemingly) sensitivity for a smooth run from 120Hz upwards (small, sealed box), with the only dip between 15kHz and 20kHz. With these speakers, cymbals sound metallic, you can hear the sound of fingers picking at guitar strings, and stereo imaging is the best I've heard (I haven't heard Mega Systems, but I've heard good ones), however, the only thing that lets these speakers down is the bass. Now, trying to integrate them with my subwoofer wasn't particularly easy when I attempted, so I looked at other ways of making bass. The idea of a FR driver with some kind of bass unit has always appealed to me, as the XO point is away from the sensitive area (3kHz), so you get the advantage of a FR driver, but with the bass of a larger, 3-way system, which brings us neatly onto the woofer.
I wanted something small (less than 8"), and cheap, but still went pretty low. After looking at various 6"+ Visaton woofers, I arrived at the W130S 8ohm, which models pretty well in a 17L box, tuned to 46Hz, and costs less than £10 each. Many of us know that cheap bass/mids often fall over on midrange if they model well (the manufacturer has to cut corners somewhere, right?), but the beauty of this design is that you can use even cheap aly drivers and avoid the ringing that is often associated with metal cones. So, bought two to make a test, make sure the go as low as winISD says, etc, and, lo and behold, they really go that low (f3 of 42Hz, f10 around 33Hz).
After discussing with Dave, it was decided that having two woofers push push, one on either side of the cabinet would be a good thing, as it would stop the whole speaker vibrating (a major thing for me, as I'm 15, so parents to please...), so, the (nearly) final design is as follows: 2x FRS8 and 2x W130S per cabinet, using a 20:1 MTL for the FR drivers, as the impedance spike at resonance is just one octave down on the XO point, and a 35.4L box (extensively braced) for the woofers.
Made one test box (just one woofer), and, after messing with phase issues etc, got a really nice sound out of it - better than expected. So, I need to make the other test box, then persuade Dad that they sound better than the Missions he lets me borrow (the old 760is), so he'll go and buy the other two.
The final cost for this design is less than £150, so it's ideal for people who are new to this, and those who want to try a new kind of speaker, without spending a fortune.
Thanks for reading, pics attached - please note - this is a test box, made of an old 3-way speaker. The cut outs are very rough, so no nasty comments on workmanship, please
http://www.diyaudio.com/forums/showthread.php?t=144099
But I decided, instead of putting my project completely onto that thread, I should start my own.
Anyway, onto the project.
Seems appropriate that I start woth driver choice, so I will.
The FR driver is a Visaton FRS8 - these are (in terms of price to performance), fantastic speakers, which sacrifice (seemingly) sensitivity for a smooth run from 120Hz upwards (small, sealed box), with the only dip between 15kHz and 20kHz. With these speakers, cymbals sound metallic, you can hear the sound of fingers picking at guitar strings, and stereo imaging is the best I've heard (I haven't heard Mega Systems, but I've heard good ones), however, the only thing that lets these speakers down is the bass. Now, trying to integrate them with my subwoofer wasn't particularly easy when I attempted, so I looked at other ways of making bass. The idea of a FR driver with some kind of bass unit has always appealed to me, as the XO point is away from the sensitive area (3kHz), so you get the advantage of a FR driver, but with the bass of a larger, 3-way system, which brings us neatly onto the woofer.
I wanted something small (less than 8"), and cheap, but still went pretty low. After looking at various 6"+ Visaton woofers, I arrived at the W130S 8ohm, which models pretty well in a 17L box, tuned to 46Hz, and costs less than £10 each. Many of us know that cheap bass/mids often fall over on midrange if they model well (the manufacturer has to cut corners somewhere, right?), but the beauty of this design is that you can use even cheap aly drivers and avoid the ringing that is often associated with metal cones. So, bought two to make a test, make sure the go as low as winISD says, etc, and, lo and behold, they really go that low (f3 of 42Hz, f10 around 33Hz).
After discussing with Dave, it was decided that having two woofers push push, one on either side of the cabinet would be a good thing, as it would stop the whole speaker vibrating (a major thing for me, as I'm 15, so parents to please...), so, the (nearly) final design is as follows: 2x FRS8 and 2x W130S per cabinet, using a 20:1 MTL for the FR drivers, as the impedance spike at resonance is just one octave down on the XO point, and a 35.4L box (extensively braced) for the woofers.
Made one test box (just one woofer), and, after messing with phase issues etc, got a really nice sound out of it - better than expected. So, I need to make the other test box, then persuade Dad that they sound better than the Missions he lets me borrow (the old 760is), so he'll go and buy the other two.
The final cost for this design is less than £150, so it's ideal for people who are new to this, and those who want to try a new kind of speaker, without spending a fortune.
Thanks for reading, pics attached - please note - this is a test box, made of an old 3-way speaker. The cut outs are very rough, so no nasty comments on workmanship, please
Attachments
Last edited:
Last night (after posting the above), I made the other test box aswell, set them up with a half-decent 50W amp, plugged into a CD player.
For the driver size, the amount of bass and volume is surprising, but I suspect there's a reason...
Most bass drums are recorded around the 40hz region, and that's where the port is tuned. I can go to 200W around port tuning and the driver is still withing Xmax, but I suspect I'd hit thermal ratings first.
Anyway, listening to those speakers, the whole concept of a low crossover and a woofer to back up a FR driver does make sense. For those who want to try it, I'd suggest you go for it - it is hugely rewarding.
Good luck!
Chris
For the driver size, the amount of bass and volume is surprising, but I suspect there's a reason...
Most bass drums are recorded around the 40hz region, and that's where the port is tuned. I can go to 200W around port tuning and the driver is still withing Xmax, but I suspect I'd hit thermal ratings first.
Anyway, listening to those speakers, the whole concept of a low crossover and a woofer to back up a FR driver does make sense. For those who want to try it, I'd suggest you go for it - it is hugely rewarding.
Good luck!
Chris
It looks like the FR driver will be in a 20:1 MTL eventually. Will both drivers be in the same space or have their own, Chris?
The build looks great, I love the cost effective approach to your experimentation, surely Dad will agree and spring for the rest of the project.
What is your next project gonna be?
I just came to ask a question on the effect of the impedance peak, but I'll answer yours first.
1) Yes, the pressure from the woofer does move the FR drivers a little, but the Qms is high enough to keep that minimal.
2) The FR drivers will probably be in the same space (Dave used a 10:1 for one driver, so 20:1 for both sounds about right), as creating two MTLs one on top of the other would result in a more difficult build...
3) After this, who knows? I'm looking towards something that will give 2x100W of clean power (>1%THD at 100W kinda thing), that still sounds good. Apart from that, possibly a centre channel, maybe a pair of rear effects speakers (currently using a pair of cheapy bookshelf speakers). Talking of which, I have some Wharfedale speakers lying around........ Back on topic.
My question is this - will the 30 ohm speak at 150Hz affect the crossover a huge amount? I could make an MTL, but if the effect of it will be minimal, I could keep a small, sealed box instead. From what I can see on winISD, the impedance peak makes the roll-off steeper lower than the peak. This wouldn't be a particularly bad thing, as far as I can see.
Chris
1) Yes, the pressure from the woofer does move the FR drivers a little, but the Qms is high enough to keep that minimal.
2) The FR drivers will probably be in the same space (Dave used a 10:1 for one driver, so 20:1 for both sounds about right), as creating two MTLs one on top of the other would result in a more difficult build...
3) After this, who knows? I'm looking towards something that will give 2x100W of clean power (>1%THD at 100W kinda thing), that still sounds good. Apart from that, possibly a centre channel, maybe a pair of rear effects speakers (currently using a pair of cheapy bookshelf speakers). Talking of which, I have some Wharfedale speakers lying around........ Back on topic.
My question is this - will the 30 ohm speak at 150Hz affect the crossover a huge amount? I could make an MTL, but if the effect of it will be minimal, I could keep a small, sealed box instead. From what I can see on winISD, the impedance peak makes the roll-off steeper lower than the peak. This wouldn't be a particularly bad thing, as far as I can see.
Chris
You are 15 years old, don't build the simplest enclosure, those are for old folks with miserable jobs and a family to support so there is no time for anything else. If the FR pair will work fine together in one MTL then thats what I'd do but if it would be better with each in its own then embrace the challenge, doesn't sound like you are skeerd to use your imagination. You might could build one 20:1 MTL enclosure with the option of installing a panel inbetween so as to experiment with each in its own then you'd know for sure.
Yes, the impedance peak will affect the XO. As can be seen from the impedance plot of FF85 in free air vrs the "aperiodic" midrange TL loading in Tysen, the box has a dramatic effect on impedance (note that no optimization has been done on the line since we went active)
dave
Actually, even if that were the case, I think what Bob is concerned about would be IM distortion from the back wave bleeding through the FR cones due to their open back. If all drivers were running full range or on an open baffle, this probably would be much less of a concern.
FRS8 is really good FR, I have it in Needles.
Why you choice to use 2 FR per channel? I reed somewhere, sound from one point is preferable, and FRS8 is enough "loud", especially with bass support.
BTW, I am thinking about Tysen style with inexpensive Visaton bass too. Maybe with this W170S:
http://www.visaton.de/en/chassis_zubehoer/tiefton/w170s_8.html
How large closed box it require? Dave, Chrisb ?
Why you choice to use 2 FR per channel? I reed somewhere, sound from one point is preferable, and FRS8 is enough "loud", especially with bass support.
BTW, I am thinking about Tysen style with inexpensive Visaton bass too. Maybe with this W170S:
http://www.visaton.de/en/chassis_zubehoer/tiefton/w170s_8.html
How large closed box it require? Dave, Chrisb ?
Last edited:
The FRS8 is 82dB at 1W/1M. (8 ohm)
Parallel that and get 88dB 1W/1M (4 ohm)
I considered the W170S, but at nearly 70L per box, I decided to leave it.
The bass performance is exceptional, with a modelled f3 of 26Hz, when tuned to 28Hz. The problem with using such a small driver is this: you use excursion really quick. Getting to 40Hz is easy enough, but after that, you need big drivers and lots of travel. The W170S is a small driver with not a huge amount of travel (the +/- 10mm shown on site is actually 10mm p/p.
At 10W, you hit Xmax at 40Hz (this is above tuning - below tuning, you run out quick), which, like I mentioned earlier, is the place of the bass drum - so you'd definately have problems.
Sealed, you get 30L, with the option of an LT circuit.
Using two per side would improve things, as you'd end up with the effective cone area of a 12", which is more capable of the bass modelled.
Chris
Parallel that and get 88dB 1W/1M (4 ohm)
I considered the W170S, but at nearly 70L per box, I decided to leave it.
The bass performance is exceptional, with a modelled f3 of 26Hz, when tuned to 28Hz. The problem with using such a small driver is this: you use excursion really quick. Getting to 40Hz is easy enough, but after that, you need big drivers and lots of travel. The W170S is a small driver with not a huge amount of travel (the +/- 10mm shown on site is actually 10mm p/p.
At 10W, you hit Xmax at 40Hz (this is above tuning - below tuning, you run out quick), which, like I mentioned earlier, is the place of the bass drum - so you'd definately have problems.
Sealed, you get 30L, with the option of an LT circuit.
Using two per side would improve things, as you'd end up with the effective cone area of a 12", which is more capable of the bass modelled.
Chris
Done some more modelling for you.
Basically, you W130S uses a box around 1/4 the size of the W170S, and the difference is around half an octave. I believe multiplying the box size by 4 to get that bit lower isn't a good, idea, but I am limited on space. You get more SPL from two of the W130s, in a box still half the size (and more power handling).
Chris
PS - am I right in thinking that if I point the speakers directly at my listening position, it will help to eliminate the comb filtering which is usually mentioned at this point.
Basically, you W130S uses a box around 1/4 the size of the W170S, and the difference is around half an octave. I believe multiplying the box size by 4 to get that bit lower isn't a good, idea, but I am limited on space. You get more SPL from two of the W130s, in a box still half the size (and more power handling).
Chris
PS - am I right in thinking that if I point the speakers directly at my listening position, it will help to eliminate the comb filtering which is usually mentioned at this point.
Hi everyone, interesting thread
Chris. Well done! Your designs are to be congratulated.
I hope you don't mind me dropping in and asking a few questions about the FRS8...
Its stated manuf spec is Qts 1.04 so I'm wondering if it is an Open Baffle candidate?
Has anyone tested its T/S parameters? Published specs may be shonky.
It looks like a great FR driver for the the price, but is it a tad ragged sounding
after 9kHz and is the 10kHz peak audible?
Grant .. I really appreciate any replies..
Chris. Well done! Your designs are to be congratulated.
I hope you don't mind me dropping in and asking a few questions about the FRS8...
Its stated manuf spec is Qts 1.04 so I'm wondering if it is an Open Baffle candidate?
Has anyone tested its T/S parameters? Published specs may be shonky.
It looks like a great FR driver for the the price, but is it a tad ragged sounding
after 9kHz and is the 10kHz peak audible?
Grant .. I really appreciate any replies..
I think the closer you can mount them to each other the better or maybe try the "Calhoun" approach and mount one on the front baffle and the other on the side. http://www.frugal-horn.com/spawn-plans.html
What do the FRS8 cost? As I remember it, the CSS SDX7 was the expensive part of the Tysen component list.
I'd prefer to keep them front mounted, and yes, the sepakers are as close together as they reasonably can be (7 or 8mm between them).
The price of this driver is £9, and the paremeters certainly match up (used winISD, let it auto-calculate), and yes, the Qts is high enough for OB use. I tried several different baffle sizes, but in the dn I ended up with one FR on an OB, one sealed behind it, to cancel the rearwaves.
The problem with these drivers is efficiency. While they sound good, 82dB requires a lot of oomph to get anywhere. The peak around 10kHz isn't audible, I think this is due to several factors: 1 - my ears aren't perfect at analysing sound, 2 - it's not in the sensitive hearing range.
Thanks for the interest + compliments
Chris
The price of this driver is £9, and the paremeters certainly match up (used winISD, let it auto-calculate), and yes, the Qts is high enough for OB use. I tried several different baffle sizes, but in the dn I ended up with one FR on an OB, one sealed behind it, to cancel the rearwaves.
The problem with these drivers is efficiency. While they sound good, 82dB requires a lot of oomph to get anywhere. The peak around 10kHz isn't audible, I think this is due to several factors: 1 - my ears aren't perfect at analysing sound, 2 - it's not in the sensitive hearing range.
Thanks for the interest + compliments
Chris
Found the link for the OB design I used
http://www.diyaudio.com/forums/showthread.php?t=142027&highlight=new+idea+to+eliminate
Chris
http://www.diyaudio.com/forums/showthread.php?t=142027&highlight=new+idea+to+eliminate
Chris
Found where I can get a more efficient version of the FR I'm using: http://www.visaton.com/en/chassis_zubehoer/breitband/frs8m_8.html they're a little more expensive, with a bigger impedance peak, but they are 6dB more efficient, meaning I only need one per side.
What do we think?
Chris
What do we think?
Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.