Hi,
some of you will know that I'm building my OB speakers since 2-3 years (in my thread "My open bafflejourney").
I decided to use parallel xo and it was fine.
But - after changing the drivers (EA15A -> 15OB350; TB W8-2145 -> TB W8-1808) and using a parallel xo again, I thought I could try the series xo. So I built a 1st order series XO, the 1808 was shorted with 10 mH (approximately 170 Hz crossover frequency for the woofer) and the woofer with 42 uF (approximately 400 Hz crossing point for the 1808). I have beautiful highs and mids, it's melting, but - where is the (deep) bass? It's considerably weaker than with a parallel XO. And that's even though everything looks fine in the simulation. As a test, I connected another 15OB350 on each side in parallel with the other woofer, in its own baffle and in an MTM configuration... the bass barely improved (but the sound image certainly did; it's considerably better, but that's not the point). Could it be that the series bass is weaker? Or what am I missing, why is it weaker for me? Any help would be greatly appreciated.
some of you will know that I'm building my OB speakers since 2-3 years (in my thread "My open bafflejourney").
I decided to use parallel xo and it was fine.
But - after changing the drivers (EA15A -> 15OB350; TB W8-2145 -> TB W8-1808) and using a parallel xo again, I thought I could try the series xo. So I built a 1st order series XO, the 1808 was shorted with 10 mH (approximately 170 Hz crossover frequency for the woofer) and the woofer with 42 uF (approximately 400 Hz crossing point for the 1808). I have beautiful highs and mids, it's melting, but - where is the (deep) bass? It's considerably weaker than with a parallel XO. And that's even though everything looks fine in the simulation. As a test, I connected another 15OB350 on each side in parallel with the other woofer, in its own baffle and in an MTM configuration... the bass barely improved (but the sound image certainly did; it's considerably better, but that's not the point). Could it be that the series bass is weaker? Or what am I missing, why is it weaker for me? Any help would be greatly appreciated.
Hi Perry,
many thanks, here they are. Measurements will follow tomorrow.
First the xo:
Here the simulated frequency response with one woofer and with factory data of the drivers:
Here the same with two woofers:
The factory data (frequency response) ist always too nice to be true, so I always measure the frequency response of my drivers in my baffle, straight without filters.
Following the simulated frequency responses using my measurements in the simulation...
...with one woofer:
...and with two woofers:
So, the difference is huge, as expected. There is a roll off below 85-90 Hz, and I think the measurement will show the same frequency response...
Here the simulated Z-graphs, with one and with two woofers:
My feeling tells me, that the measured frequency response of the woofer doesn´t look pretty - maybe a baffle issue? But I had no issue with the EA15A/W8-2145 combo... so, I'm stuck.
Cheers
many thanks, here they are. Measurements will follow tomorrow.
First the xo:
Here the simulated frequency response with one woofer and with factory data of the drivers:
Here the same with two woofers:
The factory data (frequency response) ist always too nice to be true, so I always measure the frequency response of my drivers in my baffle, straight without filters.
Following the simulated frequency responses using my measurements in the simulation...
...with one woofer:
...and with two woofers:
So, the difference is huge, as expected. There is a roll off below 85-90 Hz, and I think the measurement will show the same frequency response...
Here the simulated Z-graphs, with one and with two woofers:
My feeling tells me, that the measured frequency response of the woofer doesn´t look pretty - maybe a baffle issue? But I had no issue with the EA15A/W8-2145 combo... so, I'm stuck.
Cheers
If you want to experiment, the woofer's impedance peak at 35Hz might be helped by a resistor in series with the woofer, say 22R. With that 47uF capacitor in place that energy has nowhere to go but through the woofer.
It is a U-frame in the bass range and a simple baffle for the FR (FR don't like wings):
Half-side view, here with the 2145:
Half-side view, here with the 2145:
Well, E15A has Qts of 1.26, SB 15OB350 - 0.57.
0.57 is rather low for OB.
That woofer is barely suitable for OB, and only if heavy EQ is applied.
This was also confirmed when I simulated it with Basta a year ago.
Also in parallel configuration the series cap interacts with the voice coil inductance and produces a peak in the response, which is not the case with the series configuration.
0.57 is rather low for OB.
That woofer is barely suitable for OB, and only if heavy EQ is applied.
This was also confirmed when I simulated it with Basta a year ago.
Also in parallel configuration the series cap interacts with the voice coil inductance and produces a peak in the response, which is not the case with the series configuration.
What @Stanislav said.
Based on your photos and that woofer, you are getting the curves that I would expect. The 15A with its higher Q has about 6dB more "bump" at resonance than the 15OB350. (It just has a lot less Xmax.)
2 possible remedies:
1) Analog or DSP EQ (this is what I would recommend)
2) Passive bass boost as described here: https://www.diyaudio.com/community/...boost-4-to-7db-w-passive-xover-no-dsp.404171/
In this thread you'll see simulations of the 15OB350 for example https://www.diyaudio.com/community/...db-w-passive-xover-no-dsp.404171/post-7479682
I don't think the passive bass boost will quite get you to your goal. I suggest a 6-9dB DSP boost at 40Hz with a high pass at 35Hz to minimize excursion at low frequencies where the woofer doesn't get much traction anyway.
Based on your photos and that woofer, you are getting the curves that I would expect. The 15A with its higher Q has about 6dB more "bump" at resonance than the 15OB350. (It just has a lot less Xmax.)
2 possible remedies:
1) Analog or DSP EQ (this is what I would recommend)
2) Passive bass boost as described here: https://www.diyaudio.com/community/...boost-4-to-7db-w-passive-xover-no-dsp.404171/
In this thread you'll see simulations of the 15OB350 for example https://www.diyaudio.com/community/...db-w-passive-xover-no-dsp.404171/post-7479682
I don't think the passive bass boost will quite get you to your goal. I suggest a 6-9dB DSP boost at 40Hz with a high pass at 35Hz to minimize excursion at low frequencies where the woofer doesn't get much traction anyway.
It is strange, indeed... 🤔
My key-question would be : how many parameters did you changed in the meantime, when shifting from parallel to serial crossover configurations ?
That said, your "The Journey" Open Baffle build is very nicely made... I dig this simple, modern look.😎😎😎
T
Yes.
When you say you went to series, I should presume you response matched them. If not then the result is not likely to be about series, but to be about response.
When asking for measurements, this is one of the things I would be looking for to confirm I'm not just guessing.
Another reason to compare the old to the new is that from our point of view, something you say you think is fine becomes our frame of reference since it's difficult see through the many other factors that exist.
Well my less-than-two-cents' worth: SBA published woofer responses (including my MW19TX-8) typically showed a deep ravine at ~50hz (yours at ~45hz), then a bump up 10hz lower (yours at ~35hz). Many people dismiss the pattern as "factory measurement error" hehe. I think your 15OB350 needs at least 12-15dB boost at 45hz (sharply). Or end up 3-way like everyone else's SBA "MW".
Last edited:
Hi Stan,
yes, it is - I was a bit misled by the designation "for open baffle" – I thought it would (should) work.
The manufacturer's graph with the frequency response looks very good, but after my measurements, I had some concerns.
Hi Perry,
many thanks!
Your boost solution looks very promising, but I assume it's not (directly) usable in a Series XO. I'll play around with the solution a bit, though.
DSP - no 🙂 As you know, I'm also using high-quality analog signals, so AD/DA conversion is out of the question for me.
Back to the passive booster: why do you think it might not be a solution for me? It looks really good on paper (and with two 15OBs in parallel?).
Hi T,
you caught me. In this case, I abandoned my way of working and wanted everything at once – and changed everything. So I paid the price.
Hi Allen,
yes, I made measurements, but not with the last version, because of the sound issues - it wasn´t worth it anymore.
But here is a measurement one week ago or so, with the series xo, one 15OB woofer, 4,7mH and 42 uF.
The bass roll off is visible:
Hi,
well, it may not be a factory measurement error, but the difference is huge. I of course know how the factory mesurements will be done, but such a big difference for a woofer is remarkable:
Factory:
My measurement in baffle:
Hi,
I have to say this: the level of support in this forum is amazing. I'm always impressed. Thank you all!
I have to say this: the level of support in this forum is amazing. I'm always impressed. Thank you all!
You can likely still insert the passive bass boost circuit into your series crossover. You just need to model it and work out the kinks. It's going to be very sensitive to DCR of the inductors which needs to be low. I think it will help (4-7dB is good) I just estimate it's not quite enough boost.
