This is my first attempt at a two-way speaker. I would appreciate any constructive criticism of the design.
https://youtu.be/d4qkpSycIUg
https://youtu.be/d4qkpSycIUg
He may not understand Baffle Step, but because of the crossover frequency he chose he can simply increase the gain of the woofer and the resulting increase in bass response will counteract the baffle step response, to a first order approximation. You can see his crossover functions, with 2nd order filters I believe, around about the 7:45 mark of the video.
If, on the other hand, the crossover point was lower, e.g. around 100Hz, there would be a "hole" in the response between 200 and 500Hz that could not be filled with gain alone. So, beginner's luck on this one perhaps.
BTW, Caleb, try using a gain on the woofer that is about 1.5x-2x to boost up the bass frequencies to compensate for the baffle step losses.
Last edited:
The phase between the base and treble outputs of the x/o are matched across the frequency spectrum. And the boost in the bass signal occurs after the x/o so I think the phase will be okay.
I didn't change the gain a lot. There is a resistor divider following the high-pass filter to compensate for the higher full-range driver sensitivity. Before, it was a 1k followed by a 16.2k ohm resistor (-.5 dB). I put a 22k ohm resistor across the 16.2k to attenuate the signal to -.9 dB. It should be a pot to make it easy to adjust. If I get ambitious I'll put the whole filter/amplifier circuit on one board and include a trimpot.
Hi,
No. Bumping up the bass gain, or reducing it in the treble, changes phase addition.
rgds, sreten.
You need to get into modelling real drivers in real boxes and producing
real acoustic x/o functions like 2nd order LR acoustic with optimised
non-standard active filters designed for the specific case.
No. Bumping up the bass gain, or reducing it in the treble, changes phase addition.
rgds, sreten.
You need to get into modelling real drivers in real boxes and producing
real acoustic x/o functions like 2nd order LR acoustic with optimised
non-standard active filters designed for the specific case.
Last edited:
Today I listened to a CD using the new speakers that I've listened to in the car a number of times. There was a whole movement with low strings that I had no idea was even in the piece. I'm starting to discover aspects to music that I haven't heard in decades because I am surrounded with poor quality speakers that are meant to be barely adequate and lowest cost. And I suspect a lot of others are too.
My goal was to show people how to design decent speakers using simple tools. I think these speakers do a pretty darn good job considering the tools used to make them. If taking it to the next level requires sophisticated software, then I guess I have arrived at my destination.
My goal was to show people how to design decent speakers using simple tools. I think these speakers do a pretty darn good job considering the tools used to make them. If taking it to the next level requires sophisticated software, then I guess I have arrived at my destination.
Last edited:
Really? I don't want to threadjack but could you maybe point me to some reading or make a new thread where you elaborate on this point? Seems to me everything I've read always says that active has advantages over passive - better damping factor, no inter-modulation etc..
Speakers look cool and from what you say sounds like they sound better than what you had so congrats on the good diy project!
It's certainly not my intention to be an apologist for SReten, but I think I understand where he comes from here.
A passive crossover can't add noise, and is less likely to add distortion to the signal.
However, it's certainly hard to deny that bi-amp or tri-amp speakers have advantages including:
Direct connection between amp and driver
Less intermodulation between signals destined for different drivers
more flexibility in crossover points and slopes
Further, distortion and noise in any well-designed low level crossover are likely to be inaudible.
A passive crossover can't add noise, and is less likely to add distortion to the signal.
However, it's certainly hard to deny that bi-amp or tri-amp speakers have advantages including:
Direct connection between amp and driver
Less intermodulation between signals destined for different drivers
more flexibility in crossover points and slopes
Further, distortion and noise in any well-designed low level crossover are likely to be inaudible.
I don't think this is true. Most active speakers have a wider dynamic range than their passive counterparts.
I think i have seen the exact same speaker in it's passive form listed as 90db and the active counterpart listed as 110db.
I think meridian claimed to get a few extra dB by getting DAC as close as possible to the drivers.
All you have to do to get more distortion out of a passive crossover than an active one is to overdrive one driver in the speaker. Usually the mid driver which will run out of steam and then produce intermodulation distortions into the tweeter which would not be there if it was an active design.
Last edited:
Please explain how simple inductors, capacitors and resistors are going to add enough noise to measure.
I made no extrapolation about maximum output, but with zero input, a passively crossed-over speaker will have zero output, and therefore a signal-to-noise ration of infinity.
Let's go back to "I think I have seen", and verify before making accusations, and then apply rational thought.
"I think". Certainly anybody can make a claim. The maximum SPL of any speaker is limited by the drivers, no matter how they are driven. Manufacturers can claim whatever they want.
"I think" isn't a way to refute a statement. Use facts.
I made no extrapolation about maximum output, but with zero input, a passively crossed-over speaker will have zero output, and therefore a signal-to-noise ration of infinity.
Let's go back to "I think I have seen", and verify before making accusations, and then apply rational thought.
"I think". Certainly anybody can make a claim. The maximum SPL of any speaker is limited by the drivers, no matter how they are driven. Manufacturers can claim whatever they want.
"I think" isn't a way to refute a statement. Use facts.
I don't know electronics very well but i think they call it "passive loss".
I don't have time to find it. I was more asking you. I think it was in a Meridian article about how they felt active was better than passive. They said they had a lower noise floor and a better dynamic range than passive by getting the amps and DAC closer to the drivers. It was more something i had read - not something i am claiming is true.
I read it in a few of these sort of simplified articles released by a few manufacturers claiming that active was better than passive and then listed the reasons.
I really didn't want to jack this guys thread. I was just hoping for counter thesis to the stuff i had read that said active is far better than passive.
I think i have heard a few valid points in some of my reading where a couple of experts preferred passive over active but some of their arguments only held true because they didn't listen to music that is as aggressive as some of the stuff i listen to and there is virtually no chance of overloading and inter-modulation - not the case with the stuff i listen to.
From wikipedia - yes i know consider the source
"he main benefit of active versus passive speakers is in the higher fidelity associated with active crossovers and multiple amplifiers, including less IMD, higher dynamic range and greater output power.[1] The amplifiers within the loudspeaker enclosure may be ideally matched to the individual drivers, eliminating the need for each amplifier channel to operate in the entire audio bandpass. Driver characteristics such as power handling and impedance may be matched to amplifier capabilities.[2] More specifically, active speakers have very short speaker cables inside the enclosure, so very little voltage and control is lost in long speaker cables with higher resistance.
An active speaker often incorporates equalization tailored to each driver's response in the enclosure.[14] This yields a flatter, more neutral sound. Limiting circuits (high-ratio audio compression circuits) can be incorporated to increase the likelihood of the driver surviving high-SPL use. Such limiters may be carefully matched to driver characteristics, resulting in a more dependable loudspeaker requiring less service. Distortion detection may be designed into the electronics to help determine the onset of protective limiting, reducing output distortion and eliminating clipping.[15]"
I don't have time to find it. I was more asking you. I think it was in a Meridian article about how they felt active was better than passive. They said they had a lower noise floor and a better dynamic range than passive by getting the amps and DAC closer to the drivers. It was more something i had read - not something i am claiming is true.
I read it in a few of these sort of simplified articles released by a few manufacturers claiming that active was better than passive and then listed the reasons.
I really didn't want to jack this guys thread. I was just hoping for counter thesis to the stuff i had read that said active is far better than passive.
I think i have heard a few valid points in some of my reading where a couple of experts preferred passive over active but some of their arguments only held true because they didn't listen to music that is as aggressive as some of the stuff i listen to and there is virtually no chance of overloading and inter-modulation - not the case with the stuff i listen to.
From wikipedia - yes i know consider the source
"he main benefit of active versus passive speakers is in the higher fidelity associated with active crossovers and multiple amplifiers, including less IMD, higher dynamic range and greater output power.[1] The amplifiers within the loudspeaker enclosure may be ideally matched to the individual drivers, eliminating the need for each amplifier channel to operate in the entire audio bandpass. Driver characteristics such as power handling and impedance may be matched to amplifier capabilities.[2] More specifically, active speakers have very short speaker cables inside the enclosure, so very little voltage and control is lost in long speaker cables with higher resistance.
An active speaker often incorporates equalization tailored to each driver's response in the enclosure.[14] This yields a flatter, more neutral sound. Limiting circuits (high-ratio audio compression circuits) can be incorporated to increase the likelihood of the driver surviving high-SPL use. Such limiters may be carefully matched to driver characteristics, resulting in a more dependable loudspeaker requiring less service. Distortion detection may be designed into the electronics to help determine the onset of protective limiting, reducing output distortion and eliminating clipping.[15]"
Last edited:
Maybe this is better in a different thread?The video is cool! Nice job touching on all aspects of speaker design from a very basic perspective that will help beginners increase their understanding. Too many people just buy parts and get discouraged when it sounds bad or blows up! The learning curve is steep at first, so a comprehensive intro video like this is a good resource. Box design, active crossover, driver parameters, pretty complete. But typical beginner mistake, you did forget BSC... So did I for first several speakers. I wondered why they sounded so wimpy. Almost gave up because it seemed like black magic. Now it's still black magic concerning new topics.
Some toilet reading about BSC (baffle step correction
https://www.trueaudio.com/st_diff1.htm
Baffle Step Compensation
What these guys are talking about above, is that just boosting the volume of the woofer will not create the correct shape of the BSC needs to be.
You could simply add a separate BSC EQ circuit at the input of the crossover board, before the XO filters. The crossovers make a flat response on paper, and the BSC contours and boosts the speaker as a whole to achieve BSC. Since the box is so small, the BSC will reach up into the full range drivers band also. Use a circuit like this:
Baffle Step II (Thanks planet10)
Diffraction compensation
The center frequency of the filter is determined by the baffle width from the formulas in the trueaudio.com article, so that's a constant. Make the gain adjustable like Rod Elliot suggests so you can adjust it by ear to your preference, which will vary when it plays in different size rooms and with different distance to boundary walls which reinforce the bass to varying degree. If you have to hardwire it not adjustable then shoot for 3 or 4dB of BSC gain.
You mentioned listening to the speakers in the car. When you play low frequency sounds in a small enclosed volume like that you get a free bass boost. It is the reason car stereos can play so low with such small drivers and amps. And it's why opening the windows makes all the bass vanish! It is called cabin effect. It makes a boosting effect of 12dB per octave which is the same as the bass rolloff of a sealed speaker in a large room. So if you pick the right size woofer for the cabin, or the right size cabin for the woofer then you can achieve a flat bass response down to very low freq. So this might be why it sounds good in the car, with decent bass. But in a room the bass will sound roll off again, until you add the BSC boost. If it's destined for life in the car and you like the sound then you're done!
Have fun, and thanks for sharing your video, I enjoyed it.
Rich
So did I for first several speakers. I wondered why they sounded so wimpy. Almost gave up because it seemed like black magic. Now it's still black magic concerning new topics.Some toilet reading about BSC (baffle step correction
https://www.trueaudio.com/st_diff1.htm
Baffle Step Compensation
What these guys are talking about above, is that just boosting the volume of the woofer will not create the correct shape of the BSC needs to be.
You could simply add a separate BSC EQ circuit at the input of the crossover board, before the XO filters. The crossovers make a flat response on paper, and the BSC contours and boosts the speaker as a whole to achieve BSC. Since the box is so small, the BSC will reach up into the full range drivers band also. Use a circuit like this:
Baffle Step II (Thanks planet10)
Diffraction compensation
The center frequency of the filter is determined by the baffle width from the formulas in the trueaudio.com article, so that's a constant. Make the gain adjustable like Rod Elliot suggests so you can adjust it by ear to your preference, which will vary when it plays in different size rooms and with different distance to boundary walls which reinforce the bass to varying degree. If you have to hardwire it not adjustable then shoot for 3 or 4dB of BSC gain.
You mentioned listening to the speakers in the car. When you play low frequency sounds in a small enclosed volume like that you get a free bass boost. It is the reason car stereos can play so low with such small drivers and amps. And it's why opening the windows makes all the bass vanish! It is called cabin effect. It makes a boosting effect of 12dB per octave which is the same as the bass rolloff of a sealed speaker in a large room. So if you pick the right size woofer for the cabin, or the right size cabin for the woofer then you can achieve a flat bass response down to very low freq. So this might be why it sounds good in the car, with decent bass. But in a room the bass will sound roll off again, until you add the BSC boost. If it's destined for life in the car and you like the sound then you're done!
Have fun, and thanks for sharing your video, I enjoyed it.
Rich
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.