rrrabuyvsvsveytfazersurdwarubawvev
0% Interest for 24 Months! Learn more »
(800) 222-4700
  • Español: (800) 222-4701
Cart
June 2017 Giveaway

Sweetwater Forums [Archived]

After 15 years of great discussions, the Sweetwater Forums are now closed and preserved as a "read-only" resource. For discussions about current gear, check us out on Facebook, YouTube, inSync, and our Knowledge Base.

Y cable... Can I

goneracin

can a person use a y cord from say the mid and high outs of a crossover? Reason I ask is I have a good rane crossover, and it is actually way more advanced than our 2 way pa setup is. in 2 way mode, it will only cross the low/high at 190hz at the lowest, which is way to high for the subs. in 3 way mode, it has an adjustable sub xover from 50hz up to 300 hz or something.. BUT then there is the mid/high crossover to deal with, again lowest point being 190 hz, and highest being 7k, so I cant just not use the mid or high out, as there will still be a hole somewhere. What Id like to do if it wont harm anything, or if it will work, is to use an xlr y cable to connect the mid and high outs from the crossover, and send them to my mains amp, and obviously use the sub out to the sub amp. I know this seems like a lot of work to go through, but id like to keep this xover, as I got a very good price on it, and it is very much expandable for future use if needed.
Thanks in advance
September 6, 2008 @10:06pm
DAS

It should work okay. But before you try it you might call Rane to see if there is something that could be changed on that crossover in two-way mode to lower the crossover point. Clearly that would be better than the Y cable solution.
September 8, 2008 @04:03pm
michaelhoddy

It's a bad idea to attempt to sum outputs just by wiring them together. It MIGHT work, but you're really asking for trouble.
Which Rane crossover is it (model number)? Most of their 3-way crossovers (like the AC-23) can also operate as 2-way crossovers.
September 8, 2008 @05:18pm
dpd

^^^ Agree with Michael. If you can live with a loss in drive voltage, run each output through one end of a 10 K ohm resistor, connect the other ends of these resistors together and run that to your amp (which is probably about a 40K input impedance). that does the passive summing of the two independent outputs and keeps each opamp drive output isolated.
September 8, 2008 @10:07pm
goneracin

its an ac23b. it can opperate as a 2 way, just doesnt cross over low enough (190 hz at the lowest in 2 way mode) . I spoke with Rane today, and there are a few ways to do it for anyone interested.
1)use a line mixer to isolate the 2 outputs from each other
2)make a custom y cable with resistance in line, and across, to isolate the outputs from each other
3) buy ac22s mixer (best way) that does a better, lower 2 way crossover split.
Anyone got a Good 2 way crossover they would like to trade for a killer 3-5 way?
September 9, 2008 @04:40am
dpd

2) is easy and cheap! and it will work
September 10, 2008 @03:54am
DAS

There is a caveat to all this Y cable business I didn't mention in my earlier post (honestly, I had hoped Rane would have a better solution). Analog filters of all kinds introduce phase shift, especially around the crossover points. I don't think you can assume you can just sum together those filtered outputs and all will be as normal, as if there hadn't been drastic filtering applied. It will most certainly not sound the same as a proper two-way crossover, but as we've all agreed it will "work."
September 10, 2008 @11:45am
yeahforbes

Depending on the filter type (Linkwitz-Riley, Butterworth, etc) it may sum close to flat or there may be a dip/bump in response. What you could do to make the most of the situation is to put it where some part of the system already has a bump/dip, making it closer to flat. If it's a dip, maybe stick it where your first feedback frequency is... which is a bump in the system's response. System meaning EVERYTHING of course, from the microphone to the room's resonances.
September 10, 2008 @05:30pm
dpd

Depending on the filter type (Linkwitz-Riley, Butterworth, etc) it may sum close to flat or there may be a dip/bump in response. What you could do to make the most of the situation is to put it where some part of the system already has a bump/dip, making it closer to flat. If it's a dip, maybe stick it where your first feedback frequency is... which is a bump in the system's response. System meaning EVERYTHING of course, from the microphone to the room's resonances.

Linkwitz-Riley filters sum flat. The system gains on the drivers/amps still need to compensate for the driver efficiency differences. I know of no other crossover type that sums flat. However, the L-R filters aren't symmetrical at the crossover frequency (different rolloff rates/octave on the LP and HP sections). So, one set of drivers has to be able to respond above (below) the crossover frequency more than the other set.
September 11, 2008 @11:38pm
DAS

I don't think they are completely combinable in the time domain (phase) though. I get that they are "on paper," but group delay, etc. is an issue.
September 12, 2008 @07:35pm
dpd

^^^ Actually, the math works out perfectly in both time and frequency domains. One way to make L-R filters is to pass the signal through a filter (either HP or LP) then subtract the filtered signal from the original to get the complement filter. Thus, by definition, you get perfect summing of the electrical signals. You still need to match everything else downstream (e.g. amps, driver efficiencies and directivity) to get things exact.
Group delay is the differential of phase over frequency. So, linear phase vs frequency yields constant group delay. Conversely, integrate the group delay over frequency to get phase.
September 13, 2008 @02:48am
DAS

The math works out, but the filters themselves, in practice, don't quite. Feel free to hook it up and try it yourself.
September 15, 2008 @12:11pm
dpd

I have done both. If the hardware doesn't work, then the implementation wasn't correct - guessing gain-bandwidth would be the major contributor. Or, possibly there are some realizations that are more accurate.
September 16, 2008 @04:28am
michaelhoddy

So can he build an all-pass filter network into a Y-cable?
Yes, I'm joking.
September 16, 2008 @04:15pm
dpd

There are many ways to build all-pass filters. I know one that is really easy :D
September 17, 2008 @03:36am