Why Do Speakers Need Crossovers? How Do They Work? What's Inside Capacitors? • Bozak B-302A Gem • P2
Vložit
- čas přidán 21. 07. 2024
- In this episode, I replace the bad capacitors in this 1957 classic speaker. In the name of science and fun, I also cut the new and old capacitors open for comparison. And I explain how crossovers work, and test the Bozak’s crossover after I restore it and make some wiring changes.
Here are the topics covered in this video and the time stamps:
0:00 Intro
0:10 Quick Review of Previous Episode
0:35 Have I Even Listened to The Speaker?
0:54 It’s Got Inductors, Too!
1:05 What Does “Leaky” Capacitor Mean?
1:24 The Problem With Wax/Paper/Foil Capacitors
1:45 The New Replacement Capacitors
1:52 Capacitor Dissections
2:26 How Do The Old And New Capacitors Compare?
2:57 Metallized Film Capacitors
3:17 The Bozak Crossover N-10102 Crossover
3:21 Installing The New Capacitors
3:53 Why Do Speakers Need Crossovers?
4:08 Protecting The Tweeter
4:28 Capacitors as High Pass Filters
4:49 Protecting The Woofer
5:11 Inductors as Low Pass Filters
5:41 What If We Add a Midrange?
5:52 Band Pass Filters
6:31 How Capacitor/Inductor Values Change Cutoff Frequencies
6:44 The Bozak’s Capacitor/Inductor Values
6:58 Calculating Cutoff Frequencies
7:22 First Order Crossovers
7:29 What’s An Octave?
8:10 Comparing a 64-Year-Old Speaker to a New One
8:22 Second Order Crossovers
8:47 Third/Fourth Order Crossovers
8:55 Just Scratching The Surface
9:02 Resistors In Crossovers
9:10 Active Electronic Filters vs Passive
9:25 The Problem With Testing The Crossover
10:14 Making a Circuit Change to The Crossover
10:47 Crossover Bench Tests
10:54 Woofer Output Test
11:20 Midrange Output Test
11:32 Tweeter Output Test
To see all the videos in this series, please click:
• Bozak B-302 Gem
To visit my Amazon Shop, click here:www.amazon.com/shop/fluxconde...
#speaker #speakers #speakerrepair #hifi #1950s
Bozak Speaker. What’s inside a capacitor? How do you test a speaker crossover? How do you calculate capacitor cutoff frequency? How do you calculate inductor cutoff frequency? How do you calculate crossover values? - Věda a technologie
Loved the demonstration of the cross-over's output into real loads. Excellent teaching video!
Thanks dB, really appreciate that.
Taking the capacitors apart was fascinating. Really great video to learn from!
So cool to see the test results overlaid on the predicted curves! Excellent video.
Thanks, man! Yeah, I was really happy how well the curves matched, but also that the video effect blending the two images not only looked cool, but was also useful.
Yes he probably pissed off the other guys out there who say that doesn't work anymore it will always work they're law's of physics
Thanks for the graphics of the output in action, the test gear always interest me. Nice work.
The video I've been searching for at least a week now... Thank you.
Good, glad it helped.
Great video explaining crossovers. Hope the sound of the speakers matches the quality of your video production.
Thanks, Brian!
Awesome detailed work 👌
Excellent video and explanation of crossovers. You got a new sub and I rung that bell . Keep the videos coming. Great job!
Thanks!
Thank you for the upload, I have the same speakers and have been wondering why they make noise and sound like garbage generally, I'll definitely be cracking them open and swapping the capacitors to start and go from there
Really high quality vid. Keep it up. The subs will come
Thanks, Mike. Really appreciate that.
Excellent discussion sir!
Time for another news program too
Thanks, Martin. Yeah, I had fun doing that newscast episode and would love to do another.
Here i am Chris! Thanks for sharing it
Thanks. I really enjoy your channel.
Mind blowing.
Informative.
A good introduction to crossovers with more useful details than you often get in such videos.
However, it is impossible to get a +6dB boost in the passband of a 1st order, HPF or LPF filter into a resistive load, and this is certainly an artifact of your measurement set-up. Normally in the passband, the reactive component will approach a short curcuit and all you should get is a slight attenuation - from the cap's ESR or the Coil's copper resistance.
When connected to a real world speaker driver (which is not like a resistor!) and/or with higher order filters, you can get some narrow banded 'boosts' around the filter resonance frequency, however this is generally not acceptable, because remember there ain't no free lunch and every peak means more power pulled from the amp, at a lower and possibly harmful load impedance. Delicate speakers might also take damage from significant peaks, that's why passive filters for speakers must always only attenuate, without significant boosts, i.e. more than +1-2 dB.
The crossover tested accurately and as predicted. The methodology and design of the crossover was well detailed in the video.
@@FluxCondenser
I'm not writing this to get on your nerves by the way, it's just that I think it's important to provide correct information to especially for beginners, and the idea that a simple capacitor or inductor with boost your response by 6dB or that this would even be desirable is just not that. A passive crossover's job is to attenuate and not to boost, and any boost it can do will only be narrow banded and is generally not to be accepted for the reasons I stated before. I know it can be tricky to set the levels correctly for such measurement tools and your amp's output level here should be exactly at the 0dB reference voltage. If you can prove me wrong without a doubt. I will happily admit it, we never stop learning.
@@FluxCondenser
So *twice the voltage* is 'just as predicted' to you?
I hope the folks who build utility transformers do not think in this way, LOL !
Good Job 👌
Fantastic video love the Bozaks I own but this is why for my main system I moved to active.
Thanks, Big Papa. Glad you enjoyed the video.
@6:22 I bet you noticed already image is wrong for mid. Cap vs coil in the mid crossover blocking the frequency. What you say is obvious right.
Finally an youtube video that is a good video.
Ha! Thanks, Vale. Glad you enjoyed it.
Really well done! Your talk track is spot on!
Thanks, Andrew! Glad you enjoyed it.
Thanks for the video. I noticed you have three dummy loads... one each for the woofer, tweeter and mid. If I duplicate your setup, can I get accurate measurements with only one dummy load, moving it to the circuit path being tested? Thanks.
Bill, yes, the three sections work independently so you only need one dummy load. Just move it between the sections you want to test. I set it up with three so I could quickly switch between testing the three sections. But it’s not necessary.
Does crossover order determine the price of the speaker? For instance, am I correct in assuming that a lower distortion speaker, which is generally more cost, means it would be more suitable for 2nd, 3rd, and 4th orders? While a first order could be used on anything, it might be even better for less expensive speakers?
Some of the old capacitors are 50 + years old and still working, Im not sure if a metalized cap is going to last as long , I would think solid foil would be better
Excellent Video. As another said, may those speakers sound as good as your video production quality. Someday they will make a video about how vacuum tube biasing voltages work..for 6080 tubes. ;)
Subscribed.
Thanks, DG!
Chris. I duplicated your crossover tests and want to point out something I don't think you mentioned. While the plots are exactly as predicted using a dummy 8 Ohm load, try the same measurement with an actual driver attached. The impedance curve of the driver reeks havoc on the predicted shape of the curve. I'm sure you know this but wanted to point it out for others and see if you have any additional comments. Thanks
Thanks, Bill. Yes, this is to be expected as the speaker creates a reactive load. By reactive, we mean that the output is frequency-dependent. As frequencies change, the output goes all over the place.
This is another reason we don’t test amplifier power with a speaker and instead use a non-reactive dummy load. Wire-wound resistors aren’t suitable for this purpose as they are essentially coils which will have a reactive effect.
In your test, you’ve added a coil to the circuit with the speaker. As you saw, as the frequencies changed, so did the output, depending on the effect of the AC reactance.
By testing the crossover with a non-reactive load, we can verify that the signal going into the speaker is as-intended. But, speakers are imperfect output devices, so what happens to the signal after it goes through the voice-coil and is reproduced by the cone is subject to many variables.
Even the ability of the amplifier to supply a reactive load can affect the cone movement to a certain degree. Do a search on amplifier damping, for example. Hope this sheds a little light on a complicated subject. That’s the best I can explain it based on my knowledge.
Why does the large inductor have three leads and not two ? Does it feed two speakers ?
Nice really nice
Thanks, CB. Glad you enjoyed it.
I have listened to a lot of the full range crossover-less high end speakers. SAP, Rethym, cain and cain, wilson-benesch, Fostex and so on. In each case, they sounded boxy and midrange only. Treble was deeply rolled off, and bass wasn't great, So the claim about "audio purity" and "power robbing crossover" was lost on me. They ignore doppler distortion and rolloff.
My god! What an impressive video.A real Teacher !
Can you please help us:
I have a Bi Ampable Rainbow Reference crossover CS 260 Cross over. ( 260.30)
This has 4 terminals Bi input.
4 terminals for W- out put and 4 Terminals for T- out put.
How to connect to an amplifer ??!!
Some say Bi Input can be bridged with cables. Same we can bridge the terminals for W and T .
Don't know a b c of these!
Can u please please solve this puzzle!
Thank you.
yes
I would like to ask a question. Is it true that cheap x-over parts measure the same as high quality parts? But the sound to our ears is vastly better with the high quality parts. Some costing 10 times as much or more. What is your opinion?
My opinion is that if two components measure the same they will sound the same. Why wouldn’t they? Now, that said, there are a lot of measurements one can do on a component such as a capacitor. Far beyond just measuring for simple capacitance. And with that said, the overall circuit design of a crossover will have far greater effect on the sound than the choice of brand. Good quality capacitors, coils and resistors don’t need to be expensive. Exotic capacitors, etc are a waste of money.
Would it work with electrolytic capacitors?
Electrolyic capacitors are used in crossovers, but they need to be the non-polarized type.