mustang guy Posted June 4, 2016 Share Posted June 4, 2016 (edited) Thanks to djk for producing this epiphany for me. In a response to a post I made where I incorrectly claimed that 1.4 Ω more resistance would mean only 1/10th a dB of output difference. "1.4 Ohm difference wouldn't make more than 1/10th db difference in output." P=(v^2)/R V^2=10, R=4, P=2.5W V^2=10, R=5.4, P=1.85W 2.5W/1.85W*log*10=1.307dB (not 1/10th dB) He is spot on! A cold solder, a crusty connection from crossover screws that haven't been loosened and re-tightened, any corrosion for any reason could cause this little extra resistance on speaker wire ends or terminals. A good set of Fluke test leads will have 0.2-0.5 Ω resistance! 1.4 Ω is really not huge, and barely audible, but it could throw the overall balance of the speaker off between the different drivers in the cabinet. Overall, the output of that speaker may be just as efficient, but since it is not balanced it won't sound as good. I propose that a nice test would be to test the DC resistance at the amp side of the speaker wires to see the sum resistance of every component at one reading including: wire gauge wire quality crossover connections driver connections voice coils solder efficacy The differing results can indicate the exact loss in db as outlined in djk's calculations above. Those results can also indicate problem areas which need further research and repair. Edited June 4, 2016 by mustang guy Quote Link to comment Share on other sites More sharing options...
pzannucci Posted June 5, 2016 Share Posted June 5, 2016 You will have some issues because through the speaker wires, it is impedance not resistance. With caps and inductors in the mix, DC resistance is likely not going to be anything you really think it would be. Also balance issues could be through certain frequencies. It becomes much more helpful to measure the individual components and drivers. Make everything match and be done.... you wish. You hope that the drivers outputs measure the same. Quote Link to comment Share on other sites More sharing options...
mustang guy Posted June 5, 2016 Author Share Posted June 5, 2016 I have yet to experiment with this so I have no idea what to expect TBH. It will be interesting to see how this pans out. I think what I will be doing is seeing if the DCR is equal between like speakers. If there is a variance, I would tunnel in further to see where the added resistance is occurring. I'm hoping this can be a quick way to check internal speaker connections and speaker wire to rule that out as a problem. If I used software to capture impedance curves and compare them loudspeaker to loudspeaker, that would be better for sure, but that's beyond the scope of what I was trying to do here. In my mind, this is like compression testing cylinders, where measuring impedance sweep curves would be doing leakdown detection on a motor. Quote Link to comment Share on other sites More sharing options...
pzannucci Posted June 5, 2016 Share Posted June 5, 2016 I have yet to experiment with this so I have no idea what to expect TBH. It will be interesting to see how this pans out. I think what I will be doing is seeing if the DCR is equal between like speakers. If there is a variance, I would tunnel in further to see where the added resistance is occurring. I'm hoping this can be a quick way to check internal speaker connections and speaker wire to rule that out as a problem. If I used software to capture impedance curves and compare them loudspeaker to loudspeaker, that would be better for sure, but that's beyond the scope of what I was trying to do here. In my mind, this is like compression testing cylinders, where measuring impedance sweep curves would be doing leakdown detection on a motor. You would hope but you would be surprised if you ran something like woofer tester against: - a raw driver - a crossover with dummy loads - the actual speaker with drivers hooked up You might only be off slightly on DC resistance and could be all over the place with AC based impedance. Unless you have a bi-wire or multi-wire with no straps, you are probably reading more the woofer path than anything. A fairly rudimentary crossover might yield better results. If you have a pre-built speaker by a good manufacturer, hopefully the crossover tolerances aren't huge though they can be. Quote Link to comment Share on other sites More sharing options...
mustang guy Posted June 5, 2016 Author Share Posted June 5, 2016 Good thoughts! What I would hope to see with a simple DCR test from the speaker wires through the whole speaker network and all would be exactly the same on every like speaker so long as the speaker wire lengths are the same. I will show the results of 6 LaScalas when I get back down to the shop. I may make a special trip down today just to see. My fronts have shorter speaker wires and AL-3 networks, and the rears have very long speaker wires and AA nets. I would think that is there are big differences I would look at connections first, woofer path second, and the tolerances and other 2 drivers last. Quote Link to comment Share on other sites More sharing options...
babadono Posted June 5, 2016 Share Posted June 5, 2016 MG, as pzannucci is alluding to the capacitors in mid and high freq sections of the crossovers will block any DC resistance reading from the squawker and tweeter drivers (unless the caps are shorted or leaking badly). With a DC meter you will only be reading the speaker wire + low pass inductor + woofer voice coil + connections. Quote Link to comment Share on other sites More sharing options...
mustang guy Posted June 5, 2016 Author Share Posted June 5, 2016 MG, as pzannucci is alluding to the capacitors in mid and high freq sections of the crossovers will block any DC resistance reading from the squawker and tweeter drivers (unless the caps are shorted or leaking badly). With a DC meter you will only be reading the speaker wire + low pass inductor + woofer voice coil + connections. Gotcha! Then bad caps and speaker wire itself, but not the squaker and tweeter themselves. To me, that is still a pretty good test. It might not indicate when there is a problem, but there is little likelihood of a false positive. Quote Link to comment Share on other sites More sharing options...
pzannucci Posted June 6, 2016 Share Posted June 6, 2016 MG, as pzannucci is alluding to the capacitors in mid and high freq sections of the crossovers will block any DC resistance reading from the squawker and tweeter drivers (unless the caps are shorted or leaking badly). With a DC meter you will only be reading the speaker wire + low pass inductor + woofer voice coil + connections. Exactly. I disconnected my wire to the auto-former that manages the high frequency section on my speakers (same as removing straps on others), the others and DC resistance was the same. I had to verify anyway That's why on a lot of multi-amped configurations they still say you should use a capacitor, to insure you block DC. The DC test would at least tell you if you had a serious problem like burned or shorted voice coil or something in one of the woofers. My dual Focal 13v's even with the large coil, register 3.5 ohms. That might scare someone if they actually saw that reading as to think the speaker would toast an amp. Quote Link to comment Share on other sites More sharing options...
mustang guy Posted June 6, 2016 Author Share Posted June 6, 2016 3.5 Ω is the same as a K-33's DCR. I don't think I will see 3.5 Ω at the speaker wire side, but from what we've discussed here, I expect to see 4 to 5 Ω Quote Link to comment Share on other sites More sharing options...
pzannucci Posted June 7, 2016 Share Posted June 7, 2016 3.5 Ω is the same as a K-33's DCR. I don't think I will see 3.5 Ω at the speaker wire side, but from what we've discussed here, I expect to see 4 to 5 Ω Yes you would hope to see more when you have the inductor in the path. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.