I've assembled the passive crossovers and taken some measurements. The filters are working quite well and match very closely to the predicted response curves. Unfortunately, when testing both left and right speakers I noticed the two DE250's I have are quite different above 8kHz. There's about 2.5 dB different in level. This seems to be the case for the raw and filtered responses, so at least the speakers sound subjectively the same as they did with the digital filters.
Dan
Dan
Attachments
I did some more testing on the DE250's. After swapping CD's, waveguides, and diaphragms, I was able to narrow the difference between the two drivers to the diaphragms. I ordered a couple of replacement diaphragms so hopefully the two DE250s can be made much more "identical". It'll be interesting to see which of the two drivers, if either, was the good one.
Interestingly, this is the second quality control issue to pop up. One of the two Deltalite's had the + and - terminals swapped. When I first tried listening to these speakers, I couldn't believe how expansive but diffuse the soundstage was. After comparing impulse responses of each of the drivers and pulling the woofers out of the cabinet I saw the polarity issue. Soundstaging and imaging changed to be much more conventional though still excellent.
Dan
Interestingly, this is the second quality control issue to pop up. One of the two Deltalite's had the + and - terminals swapped. When I first tried listening to these speakers, I couldn't believe how expansive but diffuse the soundstage was. After comparing impulse responses of each of the drivers and pulling the woofers out of the cabinet I saw the polarity issue. Soundstaging and imaging changed to be much more conventional though still excellent.
Dan
Hmmm...
- #2 and #3 appear to contradict each other but thinking about it the #3 testing at THX etc I doubt tested stereo versus mono, IIRC just testing mono subs at different locations with various crossover frequencies.
- What is the data for point #2? Is there any? I suspect this REALLY depends on the blending frequency.
- I have read #5 but just anecdotally nothing scientic-IS there any actual data?
The other point I didn't quote about the shape of the room is probably a BIG confounding factor. I would hazard a hypothesis that pretty much all listening research is done in good sealed rooms. In real life, most people's listening situation is with UN-sealed rooms connected into other rooms, hallways, angled ceilings low or high etc etc etc. So maybe that research is not 100% applicable to typical home situations.
Seems like an interesting area for more investigation. Personally I may run 5.0 for my next setup...which now makes me wonder if out of an AVR I can actually blend the low bass across those. Dirac has a function that *may* do that; not sure about Audyssey if it does that inherently.
I ordered some 30 ppi reticulated open cell foam and layered up foam plugs for the SEOS to see what kind of difference they make both objectively and subjectively. So far I just have some slightly off axis measurements with and without the foam plugs. The foam looks to provide fairly constant attenuation of 1.8 dB from 2k on up, at least on this axis. From 2kHz and up, response looks similar if not identical. The attached plot has a -1.8dB offset applied to line up the two responses.
It will be interesting to hear if there is any subjective difference once I've eq'd the crossovers to give the same system response with the foam plugs in place.
Dan
It will be interesting to hear if there is any subjective difference once I've eq'd the crossovers to give the same system response with the foam plugs in place.
Dan
Attachments
Here are some listening impressions so far:
At first listen, I was surprised at the effects of adding the foam plug to the SEOS given that it seemed the response was barely changed other than the 1.8dB attenuation from 2kHz and up. It does seem to have some effect on taking the edge off some of the mid treble, in the range around 4-7kHz or so. Loud trumpet lacks a slight amount of harshness I had noticed without the foam plug. I had previously tried to reduce this with the BBC dip which I extended to around 6 kHz. It seemed like I could get away with less of a dip with the foam plugs in place.
After some back to back switching between active crossovers set up to compensate for the 1.8dB attentuation of the foam, I felt that a little brilliance and transparency in the upper treble might be missing with the foam in place. Also, with the foam plugs, it seems like imaging might be a little less solid and defined. Without the plugs, these speakers had a really solid center image that made me question the need for a center channel. With the plugs, that center image might be just a little less solid.
These are just initial impressions and I think I'll need to spend quite a bit more time comparing with and without the foam back to back to get a firmer opinion. I will also take some more polar measurements to see if the response of more different off axis.
Also, thanks for Patrick Bateman for the info on building up the plugs with 0.5 foam sheet in his thread The HOMster! (or How I Learned How to Fix a Horn)
And thanks to Earl Geddes for the initial idea of filling the waveguide with reticulated open cell foam which he used in all his speakers.
Dan
At first listen, I was surprised at the effects of adding the foam plug to the SEOS given that it seemed the response was barely changed other than the 1.8dB attenuation from 2kHz and up. It does seem to have some effect on taking the edge off some of the mid treble, in the range around 4-7kHz or so. Loud trumpet lacks a slight amount of harshness I had noticed without the foam plug. I had previously tried to reduce this with the BBC dip which I extended to around 6 kHz. It seemed like I could get away with less of a dip with the foam plugs in place.
After some back to back switching between active crossovers set up to compensate for the 1.8dB attentuation of the foam, I felt that a little brilliance and transparency in the upper treble might be missing with the foam in place. Also, with the foam plugs, it seems like imaging might be a little less solid and defined. Without the plugs, these speakers had a really solid center image that made me question the need for a center channel. With the plugs, that center image might be just a little less solid.
These are just initial impressions and I think I'll need to spend quite a bit more time comparing with and without the foam back to back to get a firmer opinion. I will also take some more polar measurements to see if the response of more different off axis.
Also, thanks for Patrick Bateman for the info on building up the plugs with 0.5 foam sheet in his thread The HOMster! (or How I Learned How to Fix a Horn)
And thanks to Earl Geddes for the initial idea of filling the waveguide with reticulated open cell foam which he used in all his speakers.
Dan
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.