passive crossover building |
Post Reply | Page 12> |
Author | |
SonicXtasy
Young Croc Joined: 02 February 2008 Location: United States Status: Offline Points: 526 |
Post Options
Thanks(0)
Posted: 07 June 2014 at 2:07pm |
http://www.partsconnexion.com/resistors_mills_mrc.html
Putting together some simple 12" 2 way tops that will be loaded with Eminence Kappalite 3012HO and a B&C DE200 on an RCF H100 horn. I just finished building the cabinets (except for the texture coat and grill) and loaded the drivers and took them to a friends house to run them through his PartsExpress.com woofer tester to see if I got the tuning right. Shooting for 60 but turned out to be 54, time to grind down some PVC. Anyway, while there we ran impedance plots of both woofers and comps. I was planning on crossing over at 1.6k which is the lowest recommended for the comp. All 4 drivers were around 12 ohms at 1.6k, is that the impedance I should build my crossovers for? Was planning 18db hp and 12db lp if it matters. Also, will these resistor in the link above work for the comp attenuation? Lastly, how would I incorporate the Eminence "light bulb" for the comp? Thanks. |
|
The member formally known as vwbidder.
|
|
odc04r
Old Croc Joined: 12 July 2006 Location: Sarfampton Status: Offline Points: 5483 |
Post Options
Thanks(0)
|
Working with the impedance at crossover point is a pretty standard way of doing it, and I've had good results doing that before. Don't forget to build your l-pad values into the crossover design too - Rod Elliot has a very good article on how to calculate values at his ESP website.
Edit: There is of course nothing wrong with mixing different orders of filter, but bear in mind the phase shift you will see at the crossover point. Use of a second order filter on both drivers is a common technique because you get a phase shift of +90 and then -90 at the crossover point for low and high driver making 180 in total. This is why the second driver polarity gets flipped to maintain phase. So if you want to mix different orders then you would be sensible to evaluate the phase difference at the crossover point and design such that when combined with the phase difference due to drivers it is as close to 0 as possible. Even if you match it their could still be issues around the xover point due to different rates of change of phase as a result of different filter orders. I would recommend sticking with symmetrical filters for an easy life if this is your first design, get more complicated later! Going a bit off topic but the next step in design that I would implement would be to measure the SPL response of both drivers with a microphone and work out the transfer function combination of that and your crossover design (per driver). Then you would pick component values via least squares minimisation such that your calculated (multiply the measured SPL curves with their crossover function and sum response over both drivers) was as flat as possible, i.e. minimum RMS deviation from an average value - you could also optimise phase in exactly the same way if it was also measured. I'm fairly sure this is what the more advanced crossover design tools do - it makes sense to me. When I get around to building my next set of HiFi cabs I am going to have a go at this! Edited by odc04r - 07 June 2014 at 2:29pm |
|
snowflake
Old Croc Joined: 29 December 2004 Location: Bristol Status: Offline Points: 3118 |
Post Options
Thanks(0)
|
put the bulb in series with the compression driver and add an L-pad before the driver/bulb that gives the right attenuation. lots of different resistor combinations will work - try one that puts a similar resistance in parallel with the driver and keeps the overall resistance reasonably high. then try different filter combinations on an active LMS before building a passive version. and you can get some extra HF response by putting a small value bypass capacitor around the L-pad. you may have to look at a notch filter on the HF and a zobel on the woofer to ensure a stable impedance around the crossover point but see how bad the problem is before you try this - sometimes it is easier to change the crossover slope or frequency rather than add extra components.
|
|
kingkwak
Young Croc Joined: 08 October 2010 Location: sussex Status: Offline Points: 810 |
Post Options
Thanks(0)
|
[QUOTE=snowflake]put the bulb in series with the compression driver and add an L-pad before the driver/bulb that gives the right attenuation.The answer to the question I was going to ask tonight, Thanks.
I used a couple of cheap QTX crossovers for the pcb base of my first one. seems to be working out ok so far.
|
|
To much power? NEVER.
|
|
SonicXtasy
Young Croc Joined: 02 February 2008 Location: United States Status: Offline Points: 526 |
Post Options
Thanks(0)
|
Thanks for the input. How much of an issue is the phase shift? I noticed most pre-built Eminence crossovers are 12db lp and 18bd hp? A friend of mine has a box full of different caps and coils and I was planning on experimenting to get the right sound. Will the Partsexpress Omnimic V2 measure everything including phase shift? Or is there a better or cheaper option? I do have a DBX Priverack PA+ with its mic but I doubt it will measure what I am looking for.
http://www.parts-express.com/dayton-audio-omnimic-v2-precision-measurement-system--390-792
|
|
The member formally known as vwbidder.
|
|
IanD
Registered User Joined: 17 January 2009 Location: London Status: Offline Points: 400 |
Post Options
Thanks(0)
|
Symmetrical filters designed for resistive loads are great in theory but often not in practice.
The inductance of the LF driver usually means that with a 2nd order filter you end up with a dip below cutoff and a peak around cutoff rather than a flat response, and this combined with the mid response of the driver usually means the response is nothing like the ideal one. This is even worse with a 3rd order filter, which also means two (expensive, power-sapping) inductors in series with the driver. The different acoustic delays of the LF driver and the HF horn mean the phases don't add up as expected anyway. For best power handling you really want at least a 3rd order cutoff for the HF driver anyway. The best way to do all this is to design the electrical network including the non-resistive parts of the driver impedance and the response of the drivers, and the series resistance of the crossover coils; doing lots of optimisation on a theoretical circuit is pointless when reality is so far from the ideal case. Edited by IanD - 09 June 2014 at 11:23pm |
|
SonicXtasy
Young Croc Joined: 02 February 2008 Location: United States Status: Offline Points: 526 |
Post Options
Thanks(0)
|
I have assembled everything for the crossover "starting point" except for the resistors. Going to start with 6db attenuation for the comp which requires a 6 and 12 ohm resistor that I don't have yet. Within a week or so I plan on setting up outside at work and really listen and "tune" with different values.
|
|
The member formally known as vwbidder.
|
|
_djk_
Old Croc Joined: 23 November 2004 Location: United States Status: Offline Points: 6002 |
Post Options
Thanks(0)
|
A 12dB/18dB crossover should work well.
The L-pad will probably need to be closer to 10dB than 6dB (the woofer is closer to 97dB in its piston range based on an No=3.25%).
The low-pass on the woofer will be non-standard.
|
|
djk
|
|
TONY.A.S.S.
Old Croc Joined: 21 February 2007 Location: United Kingdom Status: Offline Points: 6878 |
Post Options
Thanks(0)
|
I still can't get my head around running the signal through a Fuse filament. I know it's predominantly been an American thing and I understand why some people want to do it. In all my years of building passives, I've never used a Fuse. What I do do of course, is use over rated components, wire thickness etc. With all the talk of wire thickness in cable, people end up putting the signal through a red hot filament that sometimes glows in the dark.
|
|
studio45
Old Croc Joined: 16 October 2007 Location: United Kingdom Status: Offline Points: 3863 |
Post Options
Thanks(0)
|
Well it's only supposed to glow briefly, just before it pops, at a point *just* before the tweeter would have popped instead - I agree that if the system ends up getting run with the thing lit up like a bulb all the time then someone's at fault but otherwise it's a cheap way of protecting your HF
|
|
Studio45 - Repairs & Building Commotion Soundsystem -Mobile PA
|
|
djtuffer
Young Croc Joined: 02 April 2006 Location: United Kingdom Status: Offline Points: 741 |
Post Options
Thanks(0)
|
I my experience a lot of the theoretical
stuff just doesn't happen in the real world when it comes to passive crossovers. As Rog said many years ago I find the best way
is to have a handful of common values of components and build it through trial
and error. Once you realise what values of components where gives what results
you know where to go to tweak. This also assumes you have some way of measuring
the response. I don't think you can do it accurately by ear alone. |
|
cravings
Old Croc Joined: 30 January 2007 Location: Ireland Status: Offline Points: 7441 |
Post Options
Thanks(0)
|
i looked at my eminence crossovers a while ago and the lamp is actually before my l-pad.. i think really it should be after the l-pad? the lamps seemed to pop very quickly in circuit as they are. (think i just have a wire link bypassing the lamps now in those crossovers.)
|
|
Post Reply | Page 12> |
Tweet |
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |