Hi Everyone.
My first post and its complicated
I am building my first set of speakers and have run into some problems. The design is a 3-way with 8", 4" and a 3/4". The drivers are crossed at 800Hz and 5000Hz. I am using a GRS 8FR-8 Full-Range 8" for base, Peerless by Tymphany D19TD-05 3/4" Poly Dome Tweeter and a Visaton M10 4" midrange.
The box is not completed yet as I still need to add filling inside and chamferr the edges of the cabinet. The midrange and tweeter speakers are isolated from the main cavity.
The quick test last night showed a few flaws in the midrange - its almost non-existant . Bass was brilliant and the high's are great and detailed without harsness. But no warm midrange and I have a feeling that the problem is between 100Hz and 500Hz . I am not sure how to fix it though...
Heres the frequency response (Don't worry about the drop off at the end of the high range - the microfone fell over... )
Any advise and help will be appreciated
Thanks.
My first post and its complicated
I am building my first set of speakers and have run into some problems. The design is a 3-way with 8", 4" and a 3/4". The drivers are crossed at 800Hz and 5000Hz. I am using a GRS 8FR-8 Full-Range 8" for base, Peerless by Tymphany D19TD-05 3/4" Poly Dome Tweeter and a Visaton M10 4" midrange.
The box is not completed yet as I still need to add filling inside and chamferr the edges of the cabinet. The midrange and tweeter speakers are isolated from the main cavity.
The quick test last night showed a few flaws in the midrange - its almost non-existant . Bass was brilliant and the high's are great and detailed without harsness. But no warm midrange and I have a feeling that the problem is between 100Hz and 500Hz . I am not sure how to fix it though...
Heres the frequency response (Don't worry about the drop off at the end of the high range - the microfone fell over... )
An externally hosted image should be here but it was not working when we last tested it.
Any advise and help will be appreciated
Thanks.
Too much baffle step compensation. Let the woofer run more freely.
Probably so ....
I see some faults in the project.
I see that the GRS 8FR-8 has a whizzer cone.
As fullrange speakers are dimensioned differently from specialized speakers, it won't make some very low bass.
One thing to do is to cut the whizzer cone.
OR
Keep it if you like and skip the midrange driver.
But I wouldn't use it for bass only for a three-way design.I wouldn't use a whizzered speaker anyway
Look for some designs using fullrange with tweeter aid. That 19 mm poly dome would do.
I see that the GRS 8FR-8 has a whizzer cone.
As fullrange speakers are dimensioned differently from specialized speakers, it won't make some very low bass.
One thing to do is to cut the whizzer cone.
OR
Keep it if you like and skip the midrange driver.
But I wouldn't use it for bass only for a three-way design.I wouldn't use a whizzered speaker anyway
Look for some designs using fullrange with tweeter aid. That 19 mm poly dome would do.
I'd like to see the crossover schematics and speakers connections, just for fun ...
I'll put it up here.
Thanks everyone.
I am new to this and have some questions...
What is Baffle step? And how do I design the cabinet for close to a wall?
Would filling the cabinet with stuffing help with fixing things?
Smooth that response by 1/6th or 1/8th octave for a more normal looking plot. The ear tends to hear about like that. That will give you a better feel for what's going on.
That big bump at the low end is strange. I'd expect it to be higher, near 100Hz with that driver - depending of course, on the size of the box.
That big bump at the low end is strange. I'd expect it to be higher, near 100Hz with that driver - depending of course, on the size of the box.
And how do I design the cabinet for close to a wall?
First if you are close to a wall you don't need BSC, but as you ask about BSC, it's unlikely you use it and have one thing less to fix...
Second is excessive bass boost due to "room gain" or room modes excitation. Especially the case if a flat response BR is used. Maybe you are using a BR tuned around 40hz, so that a fix might be close the port to avoid this boost.
BR can be used close to a wall but with very low Qts driver and sub tuned alignments ( under FS of the driver) to get a falling response at low freqs ( Extend Bass Shelf alignments).
In a room curve +5/+10dbs under 100hz is not necessarily bad but +20dbs is overkill.
Don't know if you attenuate your mid and your tweeter, but it looks you also get this in xcess, though the response >1khz is very nicely smooth, but too low imho...
Last edited:
Maybe if using BR would try:
1. Close BR to reduce bass bump
2. Attenuate the mid or investigate the reason of this bump around crossing with the woofer.
3. Boost the tweet ( less attenuation)
Different target might be tried but all depend on individual drivers response, preferently in situ and without any xover to make an idea of what can be achieved and what not.
1. Close BR to reduce bass bump
2. Attenuate the mid or investigate the reason of this bump around crossing with the woofer.
3. Boost the tweet ( less attenuation)
Different target might be tried but all depend on individual drivers response, preferently in situ and without any xover to make an idea of what can be achieved and what not.
Attachments
Last edited:
Btw i can hardly believe that GRS have a low Qts of 0.4.
Would rather expect 0,6 or even 0,8 from this cheapo driver, hence hardly suitable for BR, hence more bump.
Suppose the datasheet simply pretends to give politically correct data for a driver apt for BR which is not, as happens in too many cases...
Would rather expect 0,6 or even 0,8 from this cheapo driver, hence hardly suitable for BR, hence more bump.
Suppose the datasheet simply pretends to give politically correct data for a driver apt for BR which is not, as happens in too many cases...
The Visaton M10 is a closed cone driver with a huge resonance at 650 Hz (Qts>4). The crossover has to take care of that, otherwise it will not sound pleasant.
OMG...
An externally hosted image should be here but it was not working when we last tested it.
Yes indeed. And as it looks like you are using HOLMImpulse, you can lock the timing of all measurements to one driver, perhaps the midrange.Separate driver responses would tell us more to understand the situation a little better.
how do I design the cabinet for close to a wall?
Neither is quite that straightforward, but chamfering is a good start. Designing with a wall is good, away from a wall can be done too, but close to a wall is more tricky.Baffle step is easy to fix. I just need to chamfer the edges of the cabinet
No, it fixes diffraction from the edges of the front plate of the box.Baffle step is easy to fix. I just need to chamfer the edges of the cabinet
Baffle step diffraction loss is very different thing, it requires careful crossover (low-pass filter) design.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Loudspeakers
- Multi-Way
- Frequency response problem