The other potential problem which is touched upon there is the low load impedance to the source with the high pass filter which can cause roll off at high frequencies, most modern sources won't have a problem with that but if you are adding a pot for level matching then it would be best to use a buffer after the pot. Then you are halfway to an active filter........
I had thought about an active filter. I am not a fan of op amps though. They are... finicky. "Close enough" engineering doesn't really work with them as they tend to amplify the difference. Also I don't really want to mess with making a PCB. Nasty chemical process I'm not a fan of or an expensive long wait from a contract MFR. Big passive components are nice and easy.
Dear twoNibbles,
Before worrying about opamps, there is is whole Universe in other priorities to right first. The first priority is trying to get your filtering in order, be it passive or active, which will require in-box measuring the drivers SPL curves plus impedance curves, and then using these in a simulation tool such as XSim.
Before, or rather without that, the whole thread will lead to a lot of handwaving. Not to a solidly designed system. There is no short cut, I am afraid.
Good Luck,
Eelco
Before worrying about opamps, there is is whole Universe in other priorities to right first. The first priority is trying to get your filtering in order, be it passive or active, which will require in-box measuring the drivers SPL curves plus impedance curves, and then using these in a simulation tool such as XSim.
Before, or rather without that, the whole thread will lead to a lot of handwaving. Not to a solidly designed system. There is no short cut, I am afraid.
Good Luck,
Eelco
Any level matching will effect the filter, but this can be taken into account in the design and so long as its something which is not adjusted later you can find a work around.
As an aside, I've just finished building Doug Self's preamp with adjustable tone controls. It uses 15 dual op amps and a multitude of electrolytic capacitors in the signal path.... the kind of thing that gives some people here nightmares
As an aside, I've just finished building Doug Self's preamp with adjustable tone controls. It uses 15 dual op amps and a multitude of electrolytic capacitors in the signal path.... the kind of thing that gives some people here nightmares
Is this really necessary? Is it impossible to arrive at a decent system without a mic and a simulation software? My gut tells me that something of reasonably high quality can be achieved via a mixture of research, listening, and tuning. I have no doubt that the way you are suggesting is the correct way but this is just my first speaker project. I am starting from nowhere. With nothing. As I build projects I will gain tools and education that will aid me in operating to your level of exactness but for now I want a reasonably successful speaker system that I can listen to and enjoy while I work on making another one to do an even better job with somewhat more educated approach from the beginning. Is this such a terrible approach?
edit: My intention is not to whine at you but honestly ask. You truly don't believe that I can attain something satisfactory in this way?
Last edited:
If you were designing a system from scratch, and particularly if you were using a high level crossover and multiple drive units, then I would agree with Boden, but that is not the case. For what you want to do ie, add a subwoofer to a full range speaker (FAST) that approach is not necessary in order to achieve the improvement you're after.
I was not aware of Mr. Elliott's website but certainly I will check it out, thank you.
Because I am so steeped in ignorance I must ask these frustratingly simple questions for clarity: When you say "high level crossover" You are talking the XO that is used in the general sense here and elsewhere, yes? That is to say an XO on the output side of the amplifier. Thus "high level" XO? Adjusting my vocabulary to that used within this community is a struggle. I think my per-existing familiarity with electronics is almost a hindrance.
So you are simply stating that you think this is a pretty decent option for me?
I think I agree. It will amount to more equipment that I will have to find a place for but I like the benefits that it affords me. If I want to change the XO of these speakers in the future I don't have to disassemble the speakers just change values in the PLLXO. If I want to experiment with ALLXO (I assume the acronym holds for the active variant) then it is a simple matter of replacing the PLLXO. If I want to try a high level XO I can easily, just remove the line level xo and the high level XO wont exist inside of the speaker box. It is an ideal setup for a perpetual tinkerer like myself. The downsides will eventually become a problem for me but at that point I should know enough to go on to project number 2.
I don't like the PLLXO acronym because in my professional world a PLL is a phase locked loop which is a clever trick for manipulating the frequency of a crystal oscillator for use as a higher frequency system clock. The subject matter is close enough that the overlap seems problematic to me. Oh well I think I have a solid plan of action now with every ones help. Will probably order parts tonight!
Ok despite the NHL playoffs being in full swing I have managed to complete the initial phase of this project: building the speakers and initial PLLXO.
The box is about 12L with a 6" long flared port for ease of finishing and because I don't understand the science behind Dave's vent design yet. Because of the last issue in the list below I will be combining the enclosures in a new enclosure. I will probably purchase Dave's planset and modify tysen to suit my needs.
Problems I am currently having in the order I plan to resolve them:
-Sub Amp does not have nearly enough head room. Can't run the ff85wks at more than 30% and I am still higher on the knob of my sub amp than I would like. I am positive I am getting distortion in the bass.
-PLLXO needs, better components, level matching, and a steeper decay. Just need to do more research here in general. May decide to go to a high level
-I messed up the boxes at the very end of the build. I drew up a sort of copy of Dave's tyson with the intention of putting the drivers towards the top front of the box sides. Well one driver wound up towards the top front... the other wound up towards the top back. Describing this error without the use of expletives is difficult.
Overall thoughts. It can sound good. But decreasing the line level volume causes the lower frequencies to disappear prematurely because of the signal lost over the series resistor in the PLLXO is my guess. I have a new amp on the way. After that I will re-evaluate my crossover. I will either redo the PLLXO, go to the high side, or explore active crossovers. I am as of yet undecided although leaning towards a high level crossover because the amps I am using are extremely budget and will be upgraded eventually and one nice amp is enough to make my wallet shiver.
Aesthetically this set turned out much nicer than my first. I am very pleased.
The box is about 12L with a 6" long flared port for ease of finishing and because I don't understand the science behind Dave's vent design yet. Because of the last issue in the list below I will be combining the enclosures in a new enclosure. I will probably purchase Dave's planset and modify tysen to suit my needs.
Problems I am currently having in the order I plan to resolve them:
-Sub Amp does not have nearly enough head room. Can't run the ff85wks at more than 30% and I am still higher on the knob of my sub amp than I would like. I am positive I am getting distortion in the bass.
-PLLXO needs, better components, level matching, and a steeper decay. Just need to do more research here in general. May decide to go to a high level
-I messed up the boxes at the very end of the build. I drew up a sort of copy of Dave's tyson with the intention of putting the drivers towards the top front of the box sides. Well one driver wound up towards the top front... the other wound up towards the top back. Describing this error without the use of expletives is difficult.
Overall thoughts. It can sound good. But decreasing the line level volume causes the lower frequencies to disappear prematurely because of the signal lost over the series resistor in the PLLXO is my guess. I have a new amp on the way. After that I will re-evaluate my crossover. I will either redo the PLLXO, go to the high side, or explore active crossovers. I am as of yet undecided although leaning towards a high level crossover because the amps I am using are extremely budget and will be upgraded eventually and one nice amp is enough to make my wallet shiver.
Aesthetically this set turned out much nicer than my first. I am very pleased.
two - it almost sounds like a minidsp should be on your short list of tools for future speaker projects. As for amps, FWIW my most recent amp "build" was with a very affordable SURE 4-channel classD amp that with an appropriate SMPS cost less than $100 USD. These things sound way better than they have any right to, and are an excellent candidate for a FAST system.
The amp in question includes a buffered selectable frequency PLLXO - if I was doing this over again, I'd probably struggle through the learning curve with a minidsp myself, particularly considering the recent addition of IR remote to their product line, and the wide range of measurement applications. A very comprehensive line well targeted for the DIY crowd, methinks.
The amp in question includes a buffered selectable frequency PLLXO - if I was doing this over again, I'd probably struggle through the learning curve with a minidsp myself, particularly considering the recent addition of IR remote to their product line, and the wide range of measurement applications. A very comprehensive line well targeted for the DIY crowd, methinks.
Hi Chris, thanks! Yeah I have considered the miniDSP and it will probably be the third or 4th experiment in the crossover experimentation I am doing it would certainly entirely avoid the question I am about to ask. As for the amps I actually started this speaker building process because I wanted to design and build my own amps. I can only take one step at a time though I'm afraid. First step is to get a system that I am thoroughly pleased with the sound of. The second step is to acquire measuring equipment and analysis software and learn that. The third step is more experiments. Then I can start thinking about electronics projects.
The question I alluded to: How big of a consideration is phase shift in analog crossovers. Obviously high and low pass crossovers phase shift in opposite directions so it shouldn't be a stretch to start seeing a cancellation around the XO freq. Maybe I don't have a solid understanding of where and how fast the phase shift happens. Anyways am I crazy or is this a thing?
The question I alluded to: How big of a consideration is phase shift in analog crossovers. Obviously high and low pass crossovers phase shift in opposite directions so it shouldn't be a stretch to start seeing a cancellation around the XO freq. Maybe I don't have a solid understanding of where and how fast the phase shift happens. Anyways am I crazy or is this a thing?
I've disappeared for awhile as I've been busy going crazy.
Yes so about how crazy I've gone. I've designed, layed out, and had made my own split DC power supply and 2nd order Linkwitz-Riley Crossover PCBs. Yes a bit on down the rabbit hole but the kind of work that I enjoy is the kind where I have no clue what I'm doing. Only one problem: driven by my PC I get an enormous buzz... Through my phone the problem doesn't exist. Definitely sounds like a ground loop problem. This confuses me, however, as my crossover/power supply doesn't have a ground connection with anything but the PC itself. Even if it did PC, amps, and crossover are all on the same outlet :/. If you all had any suggestions about how to proceed please share!
There is a small buzz through my phone but I suspect this is in part due to my project box being cardboard and no attempt being made at shielding whatsoever. At respectable volumes it is undetected only when the amps are turned up does it become noticeable.
Yes so about how crazy I've gone. I've designed, layed out, and had made my own split DC power supply and 2nd order Linkwitz-Riley Crossover PCBs. Yes a bit on down the rabbit hole but the kind of work that I enjoy is the kind where I have no clue what I'm doing. Only one problem: driven by my PC I get an enormous buzz... Through my phone the problem doesn't exist. Definitely sounds like a ground loop problem. This confuses me, however, as my crossover/power supply doesn't have a ground connection with anything but the PC itself. Even if it did PC, amps, and crossover are all on the same outlet :/. If you all had any suggestions about how to proceed please share!
There is a small buzz through my phone but I suspect this is in part due to my project box being cardboard and no attempt being made at shielding whatsoever. At respectable volumes it is undetected only when the amps are turned up does it become noticeable.
Discovered this thread today and find it very informative as I am also aiming for a FAST setup somewhere in the future.
I suspect your ground loop problem is because your PC is earthed, but your DC power supply is not. As a test you can disconnect the PC earth, but you cannot risk running your PC permanently without it. I am not sure how to ground or earth your DC power supply to avoid this.
I suspect your ground loop problem is because your PC is earthed, but your DC power supply is not. As a test you can disconnect the PC earth, but you cannot risk running your PC permanently without it. I am not sure how to ground or earth your DC power supply to avoid this.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.