I will be _extremely_ interested in what you (or others who are participating in this discussion) are eventually able to develop for a first order passive crossover for these, and how you arrive at and refine the components.
Over the holidays I built a set of small desk speakers for and with my son using the Fountek FR59EXE and with some of the new small Dayton passive radiators and I was quite literally spellbound by the coherence and depth. I don't think I have ever previously found myself "going back for just one more listen" over and over just to revel in the nuance and detail and richly 3-d soundscape that I was able to hear. I'll try to write up and post separately on those. They had remarkably good bass (just the FR59EXEs with the 2" Dayton PRs in 1 liter enclosures), but for what he likes to listen to, needed some lower end augmentation, so I built a floor-placed subwoofer using Tang Band 3" "subwoofer module.".
Based on that experience I am really, really interested in the potential of the WAW/FAST methodology, and especially with phase and response characteristics associated with first order crossovers- and I am itching to build something else using those approaches.
I find myself laughing inwardly at how much I used to sneer at the 1970s low to mid budget speakers whose tweeters merely looked like (and very nearly were) only small cheap fullrange speakers- and I know that the technology of today's small drivers (and the resulting new ability we now gain to cross the small drivers very low) is in a whole different universe from those oldschool cheapo tweeters- but who'd'a'thunk that I'd now be marvelling at the sound coming out of small drivers....
I know that there is much debate and strong opinions over whether series crossovers have any significant merit (and that they can have their own challenges) but I am really interested in the idea of doing a WAW with a series first order crossover- except that I've got a ways to go on understanding and taming the interactions of something like a series crossover with real-world, non-flat driver impedance curves.
Last edited:
Reading over my own notes, I now recall that these two drivers matched perfectly in sensitivity and a 1st order passive XO might just entail a coil and a cap plus maybe a Zobel to flatten impedance. Here was the 1st order XO from miniDSP. I did say that only mild EQ was used which probably means a simple notch filter can be used. This might be very straightforward.
I got sidetracked with work and other projects but thank you very much for sharing your thoughts on thisReading over my own notes, I now recall that these two drivers matched perfectly in sensitivity and a 1st order passive XO might just entail a coil and a cap plus maybe a Zobel to flatten impedance. Here was the 1st order XO from miniDSP. I did say that only mild EQ was used which probably means a simple notch filter can be used. This might be very straightforward.
This is from back in January. I just put the FR58EX in a new test box with a 45 degree chamfer on the back of the baffle hole. With this chamfer, there is no sign of an issue at 9 khz. In fact, once you get above the room/reflections, my measurement looks just like the data sheet FR. Thought I should issue a correction and caution about giving the driver room to breathe in back.
I studied this thread, and found that the FR58EX has 82dB sensitivity, and the AC130F1 has 89dB. The thread started out about 1st order active crossover with MiniDSP, but in the later pages turns out a 1st order passive option is also possible.
My questions:
- If used without MiniDSP, is it bi-amped due to the different sensitivity rating of the drivers?
- Is the crossover done in the line level or speaker level?
I want to do passive WAW with the two drivers:
Vifa TC9FD18-08, 3.5" Full Range -- from here on 'the FR' (full range)
ACR official site -- from here on 'the woofer'
The FR is rated between ~83-84dB and the woofer is ~88dB so there is some gap as with the original design. Before I move further, please enlight me about crossover and amplification approach. Thank you so much.
I have only done an active miniDSP with this speaker. I think that it can work with a passive 1st order XO as both drivers are fairly flat response and wide band. That’s really the requirement to make that possible.
I suppose it could be done in passive line level XO and then bi amped, but that’s complicated and the speaker is not transferable between amps easily.
If you really want to use a TC9FD in a passive 1st order speaker, you should build the speaker in this thread. Just swap 10F for TC9FD and adjust the attenuation resistor to level match.
10F/8424 & RS225-8 FAST / WAW Ref Monitor
The RS225-8 is superb driver and will give excellent sound. It’s not expensive either.
Starts with this XO:
I suppose it could be done in passive line level XO and then bi amped, but that’s complicated and the speaker is not transferable between amps easily.
If you really want to use a TC9FD in a passive 1st order speaker, you should build the speaker in this thread. Just swap 10F for TC9FD and adjust the attenuation resistor to level match.
10F/8424 & RS225-8 FAST / WAW Ref Monitor
The RS225-8 is superb driver and will give excellent sound. It’s not expensive either.
Starts with this XO:
Follow the typical design process of a two way speaker - there is a whole forum devoted to this: “Multi-way Speakers”. However, the basic process is like this - this is basically everything Inknow about how to design a 2-way speaker XO from scratch:
1. Mount both drivers in the baffle and cabinet. I would put the woofer above the full range like how I did it with the 10F/RS225 FAST speaker. Alternatively, you can put the FR58EX on top in its own chamber and baffle enclosure and slide it back about 2on to 3in for a setback to achieve some time delay of the tweeter to align its voice coil location with the woofer’s voice coil location.
2. Setup the speaker and measurement microphone on axis with the woofer (use 0.5m distance to avoid room effects) and mount the FR58EX below (off axis). This is because we will go for a transient perfect 1st order XO and FR58 needs delay. These drivers are perfect for it as they are wideband and great for having a nice step response that is time aligned.
3. Without touching the speaker position or placement of the mic placement, measure 3 sweeps at 2.83vrms excitation driving just the woofer, then just the FR58, then both in parallel. Your mic should be calibrated with a SPL meter. Now you have an SPL absolute spectrum measurement that lets you determine sensitivity from the simulation later on.
4. Load those 3 frequency response sweeps into the software of your choice Xsim (Bwaslo) or PCD (Jeff Bagley). Adjust the speaker delay alignment until the measured woofer and tweeter response when simulated as a composite looks *exactly* like the measured woofer and tweeter response (every little wiggle or peak or dip, I mean ever little one - like a fingerprint). This is basically an acoustic interferometry excercise and precisely locates the true acoustic offset between the two drivers to sub mm accuracy.
5. Now use Xsim or PCD to make your XO. I suggest overlaying the textbook 1st order XO curve and try to first EQ (via shelf and botch filters) the woofer and tweeter to match. Should not take more than a BSC shelf, and a notch or two.
6. Then select your XO frequency - I think 700Hz might be a good place to start. Apply a simple 1st order low pass coil to the woofer - circa 3.5mH to 4.5mH for an 8ohm driver. Apply a 1st order high pass cap to the tweeter circa 56uF to 62uF for an 8ohm tweeter. Apply some attenuation of the tweeter to level match.
7. Apply some impedance flattening filters.
8. The test is to check for flat phase +/-20deg from 200Hz to 15kHz if you can manage it. Then look at simulated step response function. It should look like a right triangle.
9. Then try to calculate the combinations of practical component values that are available for purchase. Build the trial XO with simple P2P soldering. Install and test by measuring the 3 sweeps: woofer only, tweeter only, composite. Compare to your simulation. Should match almost perfectly.
10. Listen to it. If you like it - congratulations! I find that the tweeter generally needs some padding to level match with woofer. I like woofer about 1.5dB to 2dB louder at 100Hz and let tweeter slowly tilt right for the best sound balance.
Note that the sensitivity of the woofer if the controlling figure. The AC130 is about 88 dB so if we account for baffle step loss of circa 5dB, the overall system sensitivity can be no higher than 83dB if you want a balanced sound. Note how this is at odds with many commercial (non horn loaded or non-PA) speakers that lie about their sensitivity - being in the 90dB+ range etc. basic HiFi woofers are never more than 89dB to 90dB. So speaker composite sensitivity with baffle step loss is always in the low to mid 80dB range.
PA speakers are different though as they are designed for high sensitivity first and foremost and are often rated 98dB - but after baffle step, they are 93dB. Still a nice figure.
Good luck!
1. Mount both drivers in the baffle and cabinet. I would put the woofer above the full range like how I did it with the 10F/RS225 FAST speaker. Alternatively, you can put the FR58EX on top in its own chamber and baffle enclosure and slide it back about 2on to 3in for a setback to achieve some time delay of the tweeter to align its voice coil location with the woofer’s voice coil location.
2. Setup the speaker and measurement microphone on axis with the woofer (use 0.5m distance to avoid room effects) and mount the FR58EX below (off axis). This is because we will go for a transient perfect 1st order XO and FR58 needs delay. These drivers are perfect for it as they are wideband and great for having a nice step response that is time aligned.
3. Without touching the speaker position or placement of the mic placement, measure 3 sweeps at 2.83vrms excitation driving just the woofer, then just the FR58, then both in parallel. Your mic should be calibrated with a SPL meter. Now you have an SPL absolute spectrum measurement that lets you determine sensitivity from the simulation later on.
4. Load those 3 frequency response sweeps into the software of your choice Xsim (Bwaslo) or PCD (Jeff Bagley). Adjust the speaker delay alignment until the measured woofer and tweeter response when simulated as a composite looks *exactly* like the measured woofer and tweeter response (every little wiggle or peak or dip, I mean ever little one - like a fingerprint). This is basically an acoustic interferometry excercise and precisely locates the true acoustic offset between the two drivers to sub mm accuracy.
5. Now use Xsim or PCD to make your XO. I suggest overlaying the textbook 1st order XO curve and try to first EQ (via shelf and botch filters) the woofer and tweeter to match. Should not take more than a BSC shelf, and a notch or two.
6. Then select your XO frequency - I think 700Hz might be a good place to start. Apply a simple 1st order low pass coil to the woofer - circa 3.5mH to 4.5mH for an 8ohm driver. Apply a 1st order high pass cap to the tweeter circa 56uF to 62uF for an 8ohm tweeter. Apply some attenuation of the tweeter to level match.
7. Apply some impedance flattening filters.
8. The test is to check for flat phase +/-20deg from 200Hz to 15kHz if you can manage it. Then look at simulated step response function. It should look like a right triangle.
9. Then try to calculate the combinations of practical component values that are available for purchase. Build the trial XO with simple P2P soldering. Install and test by measuring the 3 sweeps: woofer only, tweeter only, composite. Compare to your simulation. Should match almost perfectly.
10. Listen to it. If you like it - congratulations! I find that the tweeter generally needs some padding to level match with woofer. I like woofer about 1.5dB to 2dB louder at 100Hz and let tweeter slowly tilt right for the best sound balance.
Note that the sensitivity of the woofer if the controlling figure. The AC130 is about 88 dB so if we account for baffle step loss of circa 5dB, the overall system sensitivity can be no higher than 83dB if you want a balanced sound. Note how this is at odds with many commercial (non horn loaded or non-PA) speakers that lie about their sensitivity - being in the 90dB+ range etc. basic HiFi woofers are never more than 89dB to 90dB. So speaker composite sensitivity with baffle step loss is always in the low to mid 80dB range.
PA speakers are different though as they are designed for high sensitivity first and foremost and are often rated 98dB - but after baffle step, they are 93dB. Still a nice figure.
Good luck!
Last edited:
FR58EX and AC130F1
XRK971,
That is a very complete description. Thank you for taking the time to point me in the right direction. I will give it the old college try and see what i come up with. I am not very good at taking measurements but I do have the equipment. I have worked with PCD but never really got good results. I have finished my RS225 /10F project and I am enjoying them very much. I am inspired by all of your projects and have been doing a lot of reading. I have a miniDSP but I have no idea how to use it. Again, Thank you.
PS: Any thoughts about how to get a 3" offset? Separate box?
XRK971,
That is a very complete description. Thank you for taking the time to point me in the right direction. I will give it the old college try and see what i come up with. I am not very good at taking measurements but I do have the equipment. I have worked with PCD but never really got good results. I have finished my RS225 /10F project and I am enjoying them very much. I am inspired by all of your projects and have been doing a lot of reading. I have a miniDSP but I have no idea how to use it. Again, Thank you.
PS: Any thoughts about how to get a 3" offset? Separate box?
Last edited:
Yes, separate box is easiest. However, you have a reflection from the larger box top wall. Alternatively, a 3in deep waveguide could work too. You could 3D print one.
MiniDSP is the easiest route. If you simple get two amps (inexpensive Class D ones are fine) you can skip the whole passive XO exercise. However, there is something very visceral with passive crossovers done right. I also like how they work every time and with every amp. And only 1 amp.
I hope you do it - I can guide you through it on the thread if you wish. Another option is to make a tapered TL for the woofer to snappy solid bass response extended down to 50Hz or so.
MiniDSP is the easiest route. If you simple get two amps (inexpensive Class D ones are fine) you can skip the whole passive XO exercise. However, there is something very visceral with passive crossovers done right. I also like how they work every time and with every amp. And only 1 amp.
I hope you do it - I can guide you through it on the thread if you wish. Another option is to make a tapered TL for the woofer to snappy solid bass response extended down to 50Hz or so.
Last edited:
Ready to go
I see that you made your dagger for the FR58EX 3" x 5.75" . Is that roughly the size you would recommend for a separate box to house it ?
I have a box with the AC130 in it that is 6" x 11" x 18" (Inside dimensions). Will that work ok? Is it too big? I thought about mounting the AC130 close to the top and then put the box with the FR58 on top 2.5" from the edge.
I am just throwing out my ideas but I am open to any suggestions of yours.
I tried this woofer with a good SB tweeter but I never could make it sing for me. Again, Thank you for your help and encouragement. You might even convince me to get out my miniDSP. I have only had it about 5 years.
Howard
I see that you made your dagger for the FR58EX 3" x 5.75" . Is that roughly the size you would recommend for a separate box to house it ?
I have a box with the AC130 in it that is 6" x 11" x 18" (Inside dimensions). Will that work ok? Is it too big? I thought about mounting the AC130 close to the top and then put the box with the FR58 on top 2.5" from the edge.
I am just throwing out my ideas but I am open to any suggestions of yours.
I tried this woofer with a good SB tweeter but I never could make it sing for me. Again, Thank you for your help and encouragement. You might even convince me to get out my miniDSP. I have only had it about 5 years.
Howard
xrk971,
Yesterday I finished putting the AC130f1 in the Continuum box with a new baffle. I put the FR58EX in a baffle on the front of one of the red cones. I put the cone on the Continuum box and lined it all up. I used a 3mH inductor on the woofer and a 50uf cap on the FR58. Today I took some measurements and got an FRD file but I cannot figure out how to display it. I also cannot figure how to send it to you. I would appreciate any help you can offer. Howard
Yesterday I finished putting the AC130f1 in the Continuum box with a new baffle. I put the FR58EX in a baffle on the front of one of the red cones. I put the cone on the Continuum box and lined it all up. I used a 3mH inductor on the woofer and a 50uf cap on the FR58. Today I took some measurements and got an FRD file but I cannot figure out how to display it. I also cannot figure how to send it to you. I would appreciate any help you can offer. Howard
xrk971.
I looked thru forum questions and found out how to post a picture. The info I put up is a display of an .frd file from my measurement. I will go ahead and measure each driver separately and the two together and a near field on the woofer before I move anything. i am not experienced with measuring but will do my best. I also have an IR measurement. Howard
I looked thru forum questions and found out how to post a picture. The info I put up is a display of an .frd file from my measurement. I will go ahead and measure each driver separately and the two together and a near field on the woofer before I move anything. i am not experienced with measuring but will do my best. I also have an IR measurement. Howard
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.