Announcement

Collapse
No announcement yet.

RS100-4 x2 + RS75T-8 + OT19NC00-04 Center Channel

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • gregrueff
    replied
    Crossovers assembled:

    Click image for larger version

Name:	crossover_s.jpg
Views:	68
Size:	660.4 KB
ID:	1491514

    Leave a comment:


  • randyohoh
    replied
    That set of drivers look like they are made for each other. A great looking center channel, i hope that it sounds as good as it looks, terrific.

    Leave a comment:


  • gregrueff
    replied
    Originally posted by mdocod View Post
    I've been simulating various configurations of RS drivers, and heavily focused on the RS100 series drivers to use as my mid-range, or possible mid+high range drivers for a HT rebuild. This build log provides tons of inspiration and interesting tidbits. I'm thinking I need to take my next project to the level of getting some measurement gear to do on-baffle, in-room data collection like this. Ultimately I think that is the destination most of us DIY guys are on a journey towards, and this thread may be the inspiration to get me to finally take the plunge. I'm going to have to put up more sound damping in my server room though. I don't think I have the noise floor low enough in the house here to do a good measurement. lol.

    Thanks so much for sharing this project. I really like it!!!
    I'm glad you're inspired. I try to post the type of info I enjoy reading from others.

    Don't be intimidated by a measurement setup. Although my learning curve was large, learning to measure was a big step in my understanding. You'd be surprised what kind of meaningful data you can get from terrible setups. I have a train track across the street. My noise floor is abysmal. But I can still measure accurately. The key is understanding what is important and knowing what settings to use when and at what level.

    Cheers,

    Leave a comment:


  • mdocod
    replied
    I've been simulating various configurations of RS drivers, and heavily focused on the RS100 series drivers to use as my mid-range, or possible mid+high range drivers for a HT rebuild. This build log provides tons of inspiration and interesting tidbits. I'm thinking I need to take my next project to the level of getting some measurement gear to do on-baffle, in-room data collection like this. Ultimately I think that is the destination most of us DIY guys are on a journey towards, and this thread may be the inspiration to get me to finally take the plunge. I'm going to have to put up more sound damping in my server room though. I don't think I have the noise floor low enough in the house here to do a good measurement. lol.

    Thanks so much for sharing this project. I really like it!!!

    Leave a comment:


  • Wolf
    replied
    As usual, bravo,Greg!
    Wolf

    Leave a comment:


  • gregrueff
    replied
    Verified some more things this morning:
    1. The GEQ in my HD receiver appears to be minimum phase, and with no EQ applied the chosen crossover topology still produces deep reverse nulls. If anyone chooses to build these, I would strongly encourage you to stick to the published topology but judiciously apply your own EQ for room correction as I think the impacts are significant.
    2. The impedance hump from the OT19 tweeter is not audible to me based on the current crossover. Disconnecting all the other drivers and only listening to the OT19 with both an LCR trap flattening the impedance and without one sounds exactly the same. Even listening to James Taylor, which was a great suggestion by the way. I can see how his voice would highlight the area in question. I took some measurements to see if there was something I was missing and admittedly, the HD2 is slightly higher without the LCR trap.
    Without LCR trap:
    Click image for larger version

Name:	tweeter on axis without LCR ar 12 inches.jpg
Views:	148
Size:	95.8 KB
ID:	1490831

    With LCR trap:
    Click image for larger version

Name:	tweeter on axis with LCR ar 12 inches.jpg
Views:	132
Size:	95.6 KB
ID:	1490832

    Frequency response is affected minimally:
    Click image for larger version

Name:	tweeter LCR analysis.jpg
Views:	132
Size:	114.5 KB
ID:	1490833

    However, if anyone wanted to build these and couldn't bear the agony of knowing that the tweeter impendence was undamped, the LCR trap does not affect the phase significantly and one could be dropped in with no other adverse impacts to the speakers. A 1.0 mH coil paired with a 28 uF cap and 4.0 ohm resistor nail the hump at 950 Hz.

    Now the only thing left is to figure out coil placement on the crossover board and solder things up.

    They sound very, very nice.

    Cheers,

    Leave a comment:


  • gregrueff
    replied
    So here is the data plugged into VituixCad. The measurements all exhibit the monster 200 Hz room mode because nothing was equalized in these initial measurements.

    Click image for larger version

Name:	Power & DI.jpg
Views:	154
Size:	530.0 KB
ID:	1490807

    I'm struggling to understand what the Directivity Index is giving me that the power response is not.
    Using 2nd order slopes for everything, I expected the dispersion to be very even and I suppose this is verifying this.
    But what information is contained in the below and worth all those dang 60 measurements that the power response in PCD won't give me?

    Click image for larger version

Name:	Power & DI alone.jpg
Views:	139
Size:	76.4 KB
ID:	1490808

    Leave a comment:


  • davidroberts
    replied
    Originally posted by gregrueff View Post

    Hi David,

    We're discussing the response of the OT19 tweeter.

    It, like the other ring radiators does not use ferrofluid and as a result has a fairly large impedance spike at Fs.
    Some view this lack of ferrofluid as a positive, but what is not up for debate is that it can make crossover work difficult.

    Usually the high-pass roll-off gets a bump in the frequency response right where the resonance is, especially as you cross lower and lower, closer to Fs.
    One way to get rid of this is to put an LCR trap across the tweeter which neutralizes the impedance bump and basically flattens it. You can also use an L-pad to damp it, or as Wolf has stated you can put a large resistor across the tweeter to damp it as well.

    So we are debating whether the impedance bump that manifests as a frequency anomaly right above 1 kHz is far enough below the fundamental signal to be impactful (or audible).
    For the record, I have kept the impedance bump undamped since I am crossing fairly high (above Fs and by crossover standards) and it is already -17 dB down.
    Some have demonstrated that they can identify (hear) frequency anomalies as low as -30 dB down or even lower.

    Cheers!
    Roger that. So you are looking the impedance and not necessarily the predicted frequency response.

    It has been quite some time ago but I worked with Dennis Murphy on some cover designs for the RS series speakers that included a WMTW design. He was gracious enough to share his measurements with several other designers. His used the least shallow slopes and that what preferred by many builders.

    Leave a comment:


  • gregrueff
    replied
    Originally posted by davidroberts View Post
    Just for the sake of amateur hour, are you talking about the purple response line that appears at 2.5khz?
    Hi David,

    We're discussing the response of the OT19 tweeter.

    It, like the other ring radiators does not use ferrofluid and as a result has a fairly large impedance spike at Fs.
    Some view this lack of ferrofluid as a positive, but what is not up for debate is that it can make crossover work difficult.

    Usually the high-pass roll-off gets a bump in the frequency response right where the resonance is, especially as you cross lower and lower, closer to Fs.
    One way to get rid of this is to put an LCR trap across the tweeter which neutralizes the impedance bump and basically flattens it. You can also use an L-pad to damp it, or as Wolf has stated you can put a large resistor across the tweeter to damp it as well.

    So we are debating whether the impedance bump that manifests as a frequency anomaly right above 1 kHz is far enough below the fundamental signal to be impactful (or audible).
    For the record, I have kept the impedance bump undamped since I am crossing fairly high (above Fs and by crossover standards) and it is already -17 dB down.
    Some have demonstrated that they can identify (hear) frequency anomalies as low as -30 dB down or even lower.

    Cheers!

    Leave a comment:


  • davidroberts
    replied
    Just for the sake of amateur hour, are you talking about the purple response line that appears at 2.5khz?

    I have been looking for a center channel that would work well with my Anarchy MTM mains so I am really interested to see how this one turns out.

    Leave a comment:


  • Wolf
    replied
    Play some James Taylor, you'll know....

    Fair enough, if you're happy.
    I misread the graph, you are about 4k.
    Keep up the good work,
    Wolf

    Leave a comment:


  • gregrueff
    commented on 's reply
    Hey, thanks for hunting down my link!
    I am user error.

  • gregrueff
    replied
    Originally posted by Wolf View Post
    Having worked with the XT25sc50 and 90 before I would highly recommend placing a 30 Ohm resistor across it. This will get rid of that bump down there about -18 DB where the fs spikes and you may no longer have any nasality qualities. You'll notice it in some recordings just in case you haven't heard it yet. I know you are crossing about 2.8k and second order, and I did as well.

    Wolf
    Ha, I knew we'd be talking about this!
    I appreciate your input of course and foresaw the debate, so I actually tried the tweeter by itself with and without a full blown LCR trap. I personally could not hear a difference.
    I am crossing it LR2 closer to 4 kHz though, and in its current state the resonance bump is -17dB down.

    If I were to cross it at 2.8 kHz and throw a 30 ohm resister across the tweeter the peak is about -14 dB down.
    Keeping it in its current state and throwing in the resistor actually doesn't drop the peak appreciably, but does electrically damp and attenuate the resonant peak by about 4 ohms.

    I will continue to keep my ears open for any honking, but I'm not sure I am discerning enough. I decided to drop the LCR trap because I couldn't justify it to myself. I had no problem using the parts if I needed to though.
    Similarly I had a cap on the shunt leg of the midrange lowpass but dropped it because the value was so low (1.5 uF) and it only marginally made the phase tracking better. It just didn't make sense to keep the part for such a marginal benefit (to me at least).

    We have debated this before when I used the XT25SC40 for the first time and learned what a true undamped resonance peak looks like, lol, and I heeded your advice there with the full blown LCR trap. But in that instance I was crossing LR2 at 2 kHz where the bump wasn't too far down.

    But I respectfully think you just have better ears than me.

    Cheers,

    Leave a comment:


  • buggers
    commented on 's reply
    https://techtalk.parts-express.com/f...ini-monitor%20

  • a4eaudio
    commented on 's reply
    That link isn't working ("404 - File or directory not found" error), but the build is there just by doing a quick Google search. My guess is something is wrong with the cut-and-paste of the link.
Working...
X