Site Links
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Categories
In this Discussion
Please review the site Rules, Terms of Service, and Privacy Policy at your convenience. Rules, TOS, Privacy
Get familiar with the reaction system: Introducing the Reaction System
Comments
83 dB?
At least it's an 8 ohm driver - just buy 2! But then you're at $60 and still have deal with that step at 1200hz.
83dB is just right when you pair with an 89dB woofer. Looks like a decent little midrange at a decent price, the question is really if it's any better than the classic TC9/TG9
The step at 1200 is something we generally deal with in speaker design due to baffle effects, anyways. I think on a certain width of baffle, it might fill in below that 1200 before falling off due to step losses. I don't know, its definitely a visual improvement over the Vifa TC stuff, but I doubt it is a significant difference in performance.
I wondered about using baffle width to mitigate that. I'm just not that skilled yet to pull it off consistently - meaning dumb luck is my usual mode of operation.
1200Hz = 11.3 inches, so it might lend itself to a 3-way design using a 10" woofer.
That never seems to be work for me. The FR measurements are supposed to show the Baffle Step, but so far I have failed to identify it. Any help / examples of a well defined baffle step captured on FR?
If you measure a woofer from 1m distance, without crossover behind, you will see a roll off starting from 1000hz or so. That's the effect of the cabinet and a woofer interacting. Measure the same woofer in the near field and you will see fairly flat response (depending on the woofer).
Then you are designing the crossover, the woofer FR is intentionally bumped below 1000-1200 hz and in the near field measurement it looks this way. In the far field it becomes flat because it was compensated.
it's harder to see the baffle step loss on ungated measurements:
Getting a decent gate lets you see the transition from 2PI to 4PI space:
This was the MCM 55-2669 in a basic square cabinet - the one I brought to Indy several years ago. Gating correctly is critical for getting the correct amount of BSC in the circuit. In this case, I would design to a reference of ~82 db, as you are likely going to give up a little sensitivity to the series inductor, as well.
Something not often talked about is how gating affects what we see as the "baffle step region". Improper gating can lead to some confusion in how to deal properly with things.
Here is a measurement (GRS woofer in the Cabrini cabinet) that I could have done a better job gating:
I'll be one of them to say it: a lot of what we do that we call "voicing" is due to sloppy measurements!
Anyways, I believe a friend bought a pair of those mids and will be dropping them off at La Casa de Richards for some testing.
Thanks JR. Your measurements show what I am not seeing. I was going off Jeff B's Worksheets, which show a nice BSC loss based on the baffle dimensions, but could never match with the actual measurements and take an assumed freq for applying bsc and no way to verify post BSC in measurements. 200-300 Hz falls below my reliable gate which is 400+ hz. I also get lost in some three way with large woofers and low XO in the 300 hz region. 500+ or pref 700-800 for XO is safe region for me...
I've always had the best results with BSC adjustment if I use the full response: gated far-field + merged near field. It takes more work, but gives the most accurate result in the 100 - 200 Hz region. If you gate at 200 Hz there are only 2 data points from 200 - 400 Hz and as a result there is some smoothing in that octave. YMMV
@ani_101 - I was going to reply, then @Ed_Perkins said it perfectly.
For my first 3 way design, I agonized over woofer-mid balance for weeks - but finally realized that a combination of nearfield and farfield measurements had all the data I needed. Farfield measurements show you the baffle step loss, and nearfield measurements show no baffle step, so matching up those 2 right around 300-400 should show a pretty accurate view of what you hear.
The key is to make sure the near and far measurements are good enough to have some overlap to splice them properly, I've seen a lot of poorly done splicing in the past. The bigger the driver the more difficult this task becomes indoors.
Time window and Near field calculators included in VituixCAD Auxiliary menu can be of some assistance.