Here is the step response from B&W 800 Diamond: 
An externally hosted image should be here but it was not working when we last tested it.
Here is the step response from B&W 800 Diamond:
An externally hosted image should be here but it was not working when we last tested it.
Yes, I have studied this speaker and even listened to it at a high end audio listening room. I was not impressed by the sound relative to what I can make as a DIY'er when it comes to transients and percussive realism. To be fair, it is a great sounding speaker with regards to low distortion and huge dynamic range when driven by a good powerful amp. The Kevlar mid bass unit on B&W's is perhaps one of the best you can get and I would love to use one on my next project. You used to be able to buy these as OEM replacement parts but they require a speaker S/N now.
Grasso:
Ok, clearly stated, the XO schematic from Post #1 should have labels +ve and -ve on the speakers and both drivers are connected with +ve polarity (on top terminal shown in diagram). I will update schematic when I get back to my computer. The actual circuit as built is also connected to the driver +ve terminal in reality as verified by a +1.5vdc battery test to observe direction of motion of driver membrane (+ve = outward motion).
Regarding "thee", "thy", "thou": the affliction is worse than I originally thought as it is intentional.
I am not sure what extra specificity is gained by using the "King James" form of English is? But it's your prerogative and we will easily recognize that it's just Grasso-speak. No hard feelings - let's solve the problem at hand and we can agree to move forward. Many members have instantly recognizable language and structure and this will just be your signature.
So now that we know that all drivers are driven with regular positive polarity, let's help me solve why it's behaving so strange relative to a standard 1st order transient perfect. It should have been simple (but maybe that's my naïveté for trying to make a passive 1st order with just a "caps and coil approach"?) but what I think is happening is a strange under damped response in the woofer. Currently 24liters and moderately stuffed with fiberglass. I will measure the impedance and get box Q again to see if it had changed. I have measured it before and recall it was about Q=0.62.
Regards,
X
Last edited:
X can it happen the vertical downward tilted lobe from 1st order summing cheat your microphone because you example have mic in same plane as FF85WK, see below what i mean. If its so then either move mic to plane in middle between W/T or change polarity for hole system so tilt is upward direction.
Attachments
Last edited:
Byrtt is right, just look at the first photograph of this thread, the tweeter is shifted much too far behind the woofer. (Reversing both electrical polarities will not reverse lobing, only turning around the loudspeaker in space will, Byrrt.) Thanks for your tolerance, i just like things to have a strong logical foundation (for i believe that i can control logic).
Thanks clear my bummer you absolute right woofer will lag relative to tweeter no matter how electrical polarities is wired, post 25 graphs is then wrong below should clear things.
Attachments
Last edited:
I was going to stay out of this conversation because I am a novice at XO design, but....
You either have serious room related bass boom or too small a box (assuming sealed) for the RSS225. To make the XO work as expected, you need to EQ both drivers to flat a couple of octaves both sides of the XO point and then adjust the levels. Of course, this is easily accomplished with DSP. Passively, this is going to be a bear.
The point is, if you don't get a handle on the 10dB bass hump, things will continue to go sideways.
Bob
You either have serious room related bass boom or too small a box (assuming sealed) for the RSS225. To make the XO work as expected, you need to EQ both drivers to flat a couple of octaves both sides of the XO point and then adjust the levels. Of course, this is easily accomplished with DSP. Passively, this is going to be a bear.
The point is, if you don't get a handle on the 10dB bass hump, things will continue to go sideways.
Bob
+1 to what Bob says above.
Also, what's your impedance curve look like? IME you can only get 1st order to work at this frequency if you do full impedance compensation for both drivers. This quickly gets expensive if you do it with passive filters. Using mechanical means such as an aperiodic or stuffed TL enclosure combined with an RC Zobel has been a better solution for me.
As I said before, sometimes going with a series topology is more effective at this frequency.
If you don't do anything the freq response tends to follow the impedance curve.
Also, what's your impedance curve look like? IME you can only get 1st order to work at this frequency if you do full impedance compensation for both drivers. This quickly gets expensive if you do it with passive filters. Using mechanical means such as an aperiodic or stuffed TL enclosure combined with an RC Zobel has been a better solution for me.
As I said before, sometimes going with a series topology is more effective at this frequency.
If you don't do anything the freq response tends to follow the impedance curve.
Bob Brines:
I totally agree - and working on it now. I had the luxury of EQ'ing that flat in the past with miniDSP.
GregB:
I may have to put a vented and aperiodic stuffed Dagger on the FF85WK. It already has an RC Zobel to flatten the impedance. I may need to do something on the woofer as well. Next thing is to do an impedance sweep with XO in place and see how things look.
I totally agree - and working on it now. I had the luxury of EQ'ing that flat in the past with miniDSP.
GregB:
I may have to put a vented and aperiodic stuffed Dagger on the FF85WK. It already has an RC Zobel to flatten the impedance. I may need to do something on the woofer as well. Next thing is to do an impedance sweep with XO in place and see how things look.
Think Bob Brines and Greg B comments is right but in past it was possible do that simple 1st order one capacitor/inductor plus padding for 10F/RS225 combo as in picture-1 below, it failed in reality with mid-tweeter placed on same baffle as woofer, but later on with a fiberglass TG9 in a moveable dagger enclosure it showed to work when tweeter was offset back some smaller distance.
Haven't got X's in situation frq and impedance plot for FF85WK but taking the one from datasheet it will look like picture-2 below, only difference in XO compared picture-1 is capacitor for mid-tweeter have less value.
But there is something strange going on some of it in bass section, how come that low end bass hump that wasn't there in the past with RS225 sealed into same enclosure. In picture-3 i set up same XO schematic as X have come up with but use the frq response (frd-file) and impedance curve (zma-file) from past for RS225 and for FF85WK use factory data as in picture-2. Also have traced X's curve from post 1 which is the grey trace overlay in picture-3 so we can see the mystic low end bass hump that wasn't there in the past.
Sorry the extra work for X, on the other side very curious what is going on and what to learn from this setup.
EDIT 2.85mH inductor in picture-3 have ESR set to 1.25 ohm as X state into post 1.
Haven't got X's in situation frq and impedance plot for FF85WK but taking the one from datasheet it will look like picture-2 below, only difference in XO compared picture-1 is capacitor for mid-tweeter have less value.
But there is something strange going on some of it in bass section, how come that low end bass hump that wasn't there in the past with RS225 sealed into same enclosure. In picture-3 i set up same XO schematic as X have come up with but use the frq response (frd-file) and impedance curve (zma-file) from past for RS225 and for FF85WK use factory data as in picture-2. Also have traced X's curve from post 1 which is the grey trace overlay in picture-3 so we can see the mystic low end bass hump that wasn't there in the past.
Sorry the extra work for X, on the other side very curious what is going on and what to learn from this setup.
EDIT 2.85mH inductor in picture-3 have ESR set to 1.25 ohm as X state into post 1.
Attachments
Last edited:
Good point Byrtt.
Here is what I did previously with your help and it worked rather well I thought.
Passive 1st order XO was a 57uF cap with 5R and 5R voltage divider to attenuate mod/tweet, and a 6.2mH inductor for the low pass filter. The tweeter setback was 3.0in in this example.
http://www.diyaudio.com/forums/full-range/273524-10f-8424-rs225-8-fast-ref-monitor-50.html
Look at this nice SR:
Here is what I did previously with your help and it worked rather well I thought.
Passive 1st order XO was a 57uF cap with 5R and 5R voltage divider to attenuate mod/tweet, and a 6.2mH inductor for the low pass filter. The tweeter setback was 3.0in in this example.
http://www.diyaudio.com/forums/full-range/273524-10f-8424-rs225-8-fast-ref-monitor-50.html
Look at this nice SR:
Yes your 1st order setups was running nice so the RS225-8 / FF85WK combo ought to do better, below have stretched X/Y axis to same 1:1 relation for those two setups from past : )
Attachments
Yes, those were the days when my speakers behaved the way I expected them too. Lately it's like my lab is in the Bermuda Triangle where nothing works
Yes sad to hear and also irritating but when we find Bermuda Triangle spam culprit we can fire degaussing guns at it and get lab up running again
This has nothing to do with low frequency end of woofer but solely with phases around cross-over point. Try a smaller series choke and a Zobel network for the woofer, a resistor parallel to the series capacitor for the tweeter (contering open-baffle cancellation), less attentuation of tweeter level and reduced impedance of tweeter path (to reduce the 120 Hz peak). If baffle step is to become equalized, it must be done by another circuit, say a bigger choke parallel to a resistor, this in series with everything.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.