Psst... Ready to join TalkBass and start posting, make new friends, sell your gear, and more?  Register your free account in 30 seconds.

Hide your impedance mismatch

Discussion in 'Amps and Cabs [BG]' started by Funkengrooven, Jun 20, 2005.


  1. Funkengrooven

    Funkengrooven Turn it down? You gotta be nuts!!

    From previous posts all about mixing different impedance cabinets I decided to try isolating one cabinet with a crossover capacitor.
    The trick worked, the amp did not run hot and the isolated cabinet sounded fine. The only issue was that I had the cap value a little low and it removed a little too much bottom end from the the isolated cabinet. I used a 145 uF @ 250 volt capacitor. This resulted in a crossover point of about 180 Hz. I thought this might be ok based upon the fact that the slope of the crossover was 6db/octave.
    So I will now rig up a crossover capacitor of between 330 and 440 uf. this will make the crossover point between 60 and 80 Hz with the same 6 db/octave slope.
    Test results to follow.
    The isolation results because the amp does not "see" the load of the crossed over cabinet because the capacitor will not pass DC like a voice coil would do, effectively isolating the cabinet from the amp.
    Therefore if you have an 8 ohm 2x10 and a 4 ohm 1x15 and you want to run them together and your amp will not like 2.67 ohms then the cap is used with the 8 ohm cabinet isolating it from the amp which will then only "see" the 4 ohm cabinet.
     
  2. If the amp doesn't 'see' the speaker, it gets no power at those frequencies and you don't get the volume benefit of having two speaker cabinets.
     
  3. Interesting idea...
    But be careful... The amp DOES see the 2nd cab once you're over the crossover freq. The capacitor only blocks the dc your multimeter used to measure the ohms. So the 2nd speaker is really only invisible to your multimeter.

    The capacitor has an impedance that varies with freq, a high impedance at low freqs DOES isolate the 8 ohm cab from the amp, but far enough above the crossover freq, its the capacitor that effectively becomes "invisible", not the cab. Above that freq, the amp sees both cabs as if the capacitor wasn't there.

    So your amp is seeing topless shots of your 8 ohm speaker (meaning it can see the nice perky top parts clearly but can't see the bottom). If you set the crossover too low, and the amp sees enough of the lovely soft round bottom of the cab, it will start overheating due to too low a load. What normal red-blooded amp wouldn't???

    At a given freq for the 2nd cab/cap combo, load the amp sees = the Cap Z (Z=impedance) + spkr Z. At high freqs, the Cap Z drops near zero, and the speaker is not isolated. At low freqs the cab is bigger than the 8 ohms of the spkr, so essentially blocks the low freqs.

    Makes sense, right? If the speaker was really isolated from the amp, how does it get power from the amp to make sound? It can't be totally invisible to the amp and still work.

    You're "cap+spkr" combo is only mostly isolated from the other spkr when you're below the crossover freq.

    Caps DO block DC, but there's no DC in audio signals.

    I DO think the 250V rating on the cap should be fine. As long as your amp is less than 7,000 watts or so... :D

    But beware... the lower you set the crossover freq, the more of the frequency spectrum is spent with the 2nd cab visible to the power amp. There's not as much energy in high freqs for bass compared to lows, so set the crossover too low and you're asking for trouble. The higher you set it, the safer you are.

    Make sure the reason it didn't run hot was you didn't play that loud or that long. If it doesn't run too hot in a gig situation, then you may be ok.

    I'd be concerned about the cap causing the amp to see a really reactive load, not sure if that's really a problem, but I would be a little afraid to try it without making sure first.

    If your amp has a thermal shutdown, it should protect itself from running too low a load, so even if you pick a crossover point too low, it'll just thermal off as if you didn't have the capacitor hooked up. So if the reactive load isn't a problem, you should be relatively safe. Experiment with the freqs until you get a relatively cool amp with reasonable output from both cabs. Might work, but might still damage the amp, so try this at your own risk.

    Randy
     
  4. Oh, one more thing... bass has the predominant energy in.... the bass freqs. No surprise there. That means most of the power of the amp is at lower freqs too, with relatively little as you increase the freq.

    That means as the crossover goes down in freq, you get a rapid increase in the amount of power going to the 2nd cab, and a larger increase in heat production as the freq goes lower.

    Low load but not much signal (high freqs) isn't killing you. Low load plus HOT signal (low freqs) and you have high currents causing heat.

    Be prepared for a relatively small drop in freq to have a much larger effect on the heat produced by the amp. So drop it carefully and slowly

    Randy
     
  5. 44me

    44me

    Jun 17, 2002
    Bedford, NH USA
    You only have half of the circuit. To keep the amp load constant you have to block the highs to the low cab with an inductor. As Randy’s pointed out, amp heating will go up quite a bit when you lower the crossover frequency down to 80Hz. There’s no cheating physics.

    - John
     
  6. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    Your amp doesn't care about DC resistance and it has nothing to do with what's going on here. All you've done is to raise the impedance of the 'blocked' cab below the corner frequency of the capacitor. Yes, that raises system impedance in the lower end. It also causes a loss of response there. Make the capacitor large enough to lower the corner frequency and you'll lower the impedance as well, putting things right back where they were. The impedance load your amp sees is relative to AC current, not DC, because your amps output is AC current, not DC. I hope so, for if it did have a DC output of more than a few millivolts it would be a blown amp.

    Your experiment does have value. If all your drivers are the same then there is no point in using a capacitor in the circuit at all. But assuming that your tens don't respond to as low a frequency as your fifteen then any power being sent to them in the range below where they are operating with reasonable efficiency is being wasted as heat, while conversely that power cannot be utilized by the fifteen and your amp is struggling under the load. This is the whole point of a crossover, which maintains proper impedance loads, directs power to the drivers that can most effectively use it and keeps it out of those that can't. You've already proven to yourself the potential worth of the highpass section. A low pass section on the other as pointed out at by John completes the deal. The trick lies in determining at what frequency to place it for best effect.
     
  7. Funkengrooven

    Funkengrooven Turn it down? You gotta be nuts!!

    OK I get the picture...
    My rig is actually an 18 and 2x12...I used the 10's and 15 because that is what mostly is out there.
    So, What I need to do is figure out the right crossover point and buy the right caps and wind an inductor for that point and that will do it.
    I have the formulae..now for the experimentation. :hyper:
    The passive crossover system is a much better solution than blowing the amp. :D

    The spec on the 18 says that it doesn't do much over 1800 hz
    the 12's don't do much below 100 hz or so.

    so maybe I'll try to play around with between 90 and 160Hz??
     
  8. 4Mal

    4Mal Supporting Member

    Jun 2, 2002
    Columbia River Gorge
    uhhh, dumb ? but why don't you snag an active x-over and power amp. Use that to isolate your x-over point and then mimic that with the passive stuff ?
     
  9. Funkengrooven

    Funkengrooven Turn it down? You gotta be nuts!!

    I have extra amps and an active crossover...but..

    a) I would rather use just the SWR 550x (Max out @ 4 Ohms)

    b) I am moving the whole rig in my PTCruiser...at least until I get my 1969 Chevy pickup fixed, then that doesn't matter.
    The Cruiser has space and weight considerations and my rig is just about maxing it out.

    c) money...the caps/inductors are WAY cheaper than reconing the JBL 2241 for 8 ohms....

    Although the reconing is probably the way I should go.

    but I am fighting it...I just hate to tear apart a perfectly good speaker just to recone for a different impedance.. :bawl:
     
  10. 4Mal

    4Mal Supporting Member

    Jun 2, 2002
    Columbia River Gorge

    Nah .. I was thinking about using the cross over and power amp setup just to find the best cross over point. Then base your choice of passive componenets on what you find as the suitable cross over point from your experiment ...

    We seem to be the two postiing JBL bigot's of the moment. I'm playing with going full range using my 140 / 145 cabs and also a Bag End 2x10 or bi-amping. It's been a fun experiment and very suprisingly I'm finding that I like the 15's running up into the 800 to 1 k range. Getting down into the 250 to 125 range for the x-over point seems to rob a lot of punch. I get a great hi-fi Pattitucci kind of thing but no pant leg flapping ... I'm still in the initial impressions stage though with no end in sight for the experimentation stage.

    As a card-carrying JBL bigot, I'm blown away with the Bag End cab. It's the 2x10 with a coax tweeter. Very 'vintagy' kind of vibe to it. Great full range cab and it blends well with the JBL's. First 'modern' cab I've had that is a keeper ...

    Are your Weber 12's in a sealed box ? If so, what did you base the dimensions on ?
     
  11. Funkengrooven

    Funkengrooven Turn it down? You gotta be nuts!!

    Oh Yeah....JBL Bigot...that's Me!! :hyper: Is there any other speaker???

    OK I'm stupid. :spit: What a great Idea!!
    I'm glad one of us is thinkin....


    I could always use the pre-power setup that I have for the 18/2x12 and save the 550x for the 2x15...

    Yes, completely sealed. They actually handle the 550x by themselves. I made the 12's box to fit on top of the 18. Roughly the same H & W as an old 2x12 SG pa Cab I was using only deeper, and stuffed with Fibreglas ....See photos...
    The dimensions are approximately 16 H x 27 W x 16 D...or so Thor asked me what the Internal volume was ..I think it worked out to maybe 3.6 cuft? Thor, Do you remember???
     
  12. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    Don't do that. A hand wound air core coil with enough inductance to do the job is going to have a DCR of at least a couple of ohms, and if you've got an 8 ohm driver that means 1/4 of your power is going to be eaten up by the coil, 1/2 if it's a 4 ohm driver. Spend the fifteen bucks for a low DCR inductor.

    The 18 may have on-axis response to 1.8kHz but that's meaningless. The -6dB 30 degrees off-axis SPL is what counts, and 18's typically reach that limit around 400 Hz. In a sealed alignment your twelves probably start rolling off around 200 Hz, so that's where I'd start. If you don't have measuring equipement to see what's going on make sure you do your listening tests standing off to the side, not in front, and rotate those 12s so that they're vertical, otherwise you're comb filtering above 1kHz or so.
     
  13. throbgod13

    throbgod13

    Mar 26, 2005
    Texas
    cripes..

    it's pretty simple..



    MATCH YOUR IMPEDANCE..