Hello everybody, I have a slight problem that I really don’t know how to tackle cause I don’t have nobody to ask, more than you fellows here.
I have a pair of Sachiko speakers equipped with Fostex FE 208 E and JBL titanium tweeters. No, it is not the sigma model of the 208:s, just E.
When I had the cabinets made a couple of years ago at a local wood shop nearby I also had help from another person building me a filter “to give me a perfect sound” from those Sachiko. At that time I was not sure if I could accomplish such audio projects myself but I wanted to, and to be fair, I was also running my own company so the time was not on my side if we say so.
I have never questioned the construction before cause they sounded really lovely compared with my old speakers and still I was not 100 % convinced cause I had expected a little more difference.... double horn and nice speaker and 96 db sensitivity...
Now I have started to investigate what was really done to the speakers when they were put together. First I found out that the tweeters were not playing at all since I had a short circuit on the filter PCB so when that was resolved I was happy because the tweeters really added….. musicality. Wow…. those speakers are really something when all of it is playing.
But…..
Now when I have (again) returned to this lovely site I read about people feeding those Sachikos with tube amps of 7-15 Watts and they describe them as outstanding and of really good quality and that impressed I am not…. so, what is wrong with my pair of speakers?
This??
I have been reading about a lot of people who only use a single capacitor as high pass for tweeters in this kind of cabinet solution and then I started to think that it might be something wrong with the filters that….
Well, I have now taken one filter out and now something really happened (again), it was like lifting the lid of the kettle and down there are the musicians playing and now, suddenly I can here them. WOWW!!
The sound is more open, airy and a lot of more details, three dimensional and well, more of everything but what surprise me most is that I have regained some 1.5-2 db in sensitivity, clearly louder and since I am right now listen to one speaker, still with the filter and the other one straight through I can, without measurements say that the one without filter definitely is louder but…..
I have lost some gain on the LF so now my question is, can I regain some LF but without loosing all the good that I received by dismantling the filter from the signal path that was there before or can I build another filter that will be better. My idea right now is to just put a cap of about 1-1.5 microF for the tweeter and leave it as is.
I can’t wait til tomorrow cause then the other speaker goes to surgery to remove a filter!
Then I will also see (hear) how much it affects the LF in total, it is a little uncertain to evaluate from just one speaker. Maybe I will be satisfied anyway but I am worrying a little about the LF cause there is definitely some loss, that I know. Isn’t the Sachiko going rather deep, somewhere I recall to have heard about 30-35 Hz to be the low for those speakers.
I also want to know how important is it that the back of the magnet come to touch the construction of the compression chamber? Will it help my LF or it doesn’t matter? I have also started experimenting with some damping in the compression chamber, we will se where I end up but I am convinced that my speakers still have a lot to show.
Now when I re-experience the Sachiko for the first time and hear their real potential I realize that my next coming project will be just heaven for those speakers, Tom Christiansen’s “Damn Good Amp 300B”! My first tube amp project and most likely not my last cause I also aim for a tube pre amplifier but in time…. now I want the speakers running as they are ment to sing and then the amp…(s).
I have a pair of Sachiko speakers equipped with Fostex FE 208 E and JBL titanium tweeters. No, it is not the sigma model of the 208:s, just E.
When I had the cabinets made a couple of years ago at a local wood shop nearby I also had help from another person building me a filter “to give me a perfect sound” from those Sachiko. At that time I was not sure if I could accomplish such audio projects myself but I wanted to, and to be fair, I was also running my own company so the time was not on my side if we say so.
I have never questioned the construction before cause they sounded really lovely compared with my old speakers and still I was not 100 % convinced cause I had expected a little more difference.... double horn and nice speaker and 96 db sensitivity...
Now I have started to investigate what was really done to the speakers when they were put together. First I found out that the tweeters were not playing at all since I had a short circuit on the filter PCB so when that was resolved I was happy because the tweeters really added….. musicality. Wow…. those speakers are really something when all of it is playing.
But…..
Now when I have (again) returned to this lovely site I read about people feeding those Sachikos with tube amps of 7-15 Watts and they describe them as outstanding and of really good quality and that impressed I am not…. so, what is wrong with my pair of speakers?
This??
I have been reading about a lot of people who only use a single capacitor as high pass for tweeters in this kind of cabinet solution and then I started to think that it might be something wrong with the filters that….
Well, I have now taken one filter out and now something really happened (again), it was like lifting the lid of the kettle and down there are the musicians playing and now, suddenly I can here them. WOWW!!
The sound is more open, airy and a lot of more details, three dimensional and well, more of everything but what surprise me most is that I have regained some 1.5-2 db in sensitivity, clearly louder and since I am right now listen to one speaker, still with the filter and the other one straight through I can, without measurements say that the one without filter definitely is louder but…..
I have lost some gain on the LF so now my question is, can I regain some LF but without loosing all the good that I received by dismantling the filter from the signal path that was there before or can I build another filter that will be better. My idea right now is to just put a cap of about 1-1.5 microF for the tweeter and leave it as is.
I can’t wait til tomorrow cause then the other speaker goes to surgery to remove a filter!
Then I will also see (hear) how much it affects the LF in total, it is a little uncertain to evaluate from just one speaker. Maybe I will be satisfied anyway but I am worrying a little about the LF cause there is definitely some loss, that I know. Isn’t the Sachiko going rather deep, somewhere I recall to have heard about 30-35 Hz to be the low for those speakers.
I also want to know how important is it that the back of the magnet come to touch the construction of the compression chamber? Will it help my LF or it doesn’t matter? I have also started experimenting with some damping in the compression chamber, we will se where I end up but I am convinced that my speakers still have a lot to show.
Now when I re-experience the Sachiko for the first time and hear their real potential I realize that my next coming project will be just heaven for those speakers, Tom Christiansen’s “Damn Good Amp 300B”! My first tube amp project and most likely not my last cause I also aim for a tube pre amplifier but in time…. now I want the speakers running as they are ment to sing and then the amp…(s).
Last edited:
I'm not familiar with Sashisko. Do you mean Sachiko speakers??
If these had passive filters on both the Fostex and the tweeter, it's no surprise they didn't sound good. Who knows what your friend did. It's usually a mistake to try to get full range drivers perfectly flat with passive filters. Also, those parts look pretty questionable. There's no place for ferrite core inductors in any crossover IMO. It was a mistake to try to make this a conventional two way, if that's what was done.
Anyway, yeah, run the Fostex wide open, and if you feel you need a supertweeter, highpass it with a ~1uF cap and (optionally) an L-pad for level. Frankly you don't really need a supertweeter at all.
If these had passive filters on both the Fostex and the tweeter, it's no surprise they didn't sound good. Who knows what your friend did. It's usually a mistake to try to get full range drivers perfectly flat with passive filters. Also, those parts look pretty questionable. There's no place for ferrite core inductors in any crossover IMO. It was a mistake to try to make this a conventional two way, if that's what was done.
Anyway, yeah, run the Fostex wide open, and if you feel you need a supertweeter, highpass it with a ~1uF cap and (optionally) an L-pad for level. Frankly you don't really need a supertweeter at all.
I feel that there is need for the tweeter since the 208 E fall off already at 16000 Hz. The 208 E sigma go all up to 18500 before it start to roll off so there it is not that obvious but at 16000 it is a little to early.
But my main question is if there is a way to regain som LF without loosing on the upper side now?
But my main question is if there is a way to regain som LF without loosing on the upper side now?
Last edited:
I think you have 3 basic options:
1) Move the speakers closer to your back wall, or even into a corner.
2) Make a simple but high quality passive LR shelving filter, using a good air core or laminate core inductor and good quality power resistors. Quality really matters here. You may want to add a bypass cap too, to preserve the top octave.
3) Use a high resolution digital EQ. This should be simple enough if your source is digital, and the losses will be minimal.
If you are going to a high output impedance tube amplifier, this will increase the bass also.
Perhaps P10 or scottmoose will chime in.
WOW! I have a grin spreading from one ear to the other.
The second speaker went into surgery today and the last three hours have been spent listening to music after i put them together again.
Next step is to lower the volume of the tweeter and to apply a decent cap of sutable value. I will though cut it off rather low since it is bringing a lot of details to even a cello and counterbass(?). The lower, the better but without straining it (hope you understand my intention).
The second speaker went into surgery today and the last three hours have been spent listening to music after i put them together again.
Next step is to lower the volume of the tweeter and to apply a decent cap of sutable value. I will though cut it off rather low since it is bringing a lot of details to even a cello and counterbass(?). The lower, the better but without straining it (hope you understand my intention).
Last edited:
Hi TIC37,
Just ignore my suggetions if not a way to go you the the boss, but if you actual want to know more about where you are instead of hey this sounds entertaining, the tools are free only calibrated microphone add cost.
Free REW software to measure response of both drivers and or room, if going passive two way REW via soundcard and a cheap resistor can measure impedance of drivers sitting in their box which can help make the passive XO network more precise. Link REW - Room EQ Wizard Room Acoustics Software. Will work on Windows and Mac.
Link to knowledge around baffle step and diffraction choose topic 2 there True Audio's Loudspeaker Design Tech Topics Index.
To simulate baffle step and diffraction link to great free Swedish tool choose The Edge and download Tolvan Data. Will work on Windows.
To simulate baffle step and diffraction plus room modes link to free Excel application software and choose "Baffle Diffraction and Boundary Simulator". Will work on Windows and needs MS Excel license both old and new versions are supported.
If going two way to find best XO point try JRiver mediacenter where its DSP engine have all plugins build in, it's not meant as your future setup just to find the best XO point and slopes for you drivers that then later can be build as passive ones. In the route set up the DSP engine to mix L and R to mono then put a choosed LP XO slope to route to left output and a choosed HP XO slope to route to right output. Now your power amps left channel shall feed full ranger and right channel feed tweeter. On the fly you can change filters slope and frequency to find best ones without buying real components, listen and measure to a mono speaker and you could even add EQ if wanted. Link JRiver Media Center software. Works on Windows and Mac, although unsure if Mac has as many plugins as Windows version, the trail version have full function in a month. Tip to measure with REW and have JR engine filters looped into Windows sound system, choose the virtual sound card "JRiver Media Center 20" as default soundcard and keep JR program open in back ground, then all programs making sound will be looped through what ever settings JR DSP engine add.
Passive XO simulator to tune passive XO, very good to reach precision of a certain intended slope. Link http://www.diyaudio.com/forums/multi-way/259865-xsim-free-crossover-designer.html. Work on Windows probably need some exercise time but think is worth the time.
Calibrated microphone at 89€ a German webshop, USB interface works on Windows and Mac. Link miniDSP USB Messmikrofon UMIK-1 | Lautsprcher Selbstbau.
Hope links not dead or pointing wrong 🙂.
Just ignore my suggetions if not a way to go you the the boss, but if you actual want to know more about where you are instead of hey this sounds entertaining, the tools are free only calibrated microphone add cost.
Free REW software to measure response of both drivers and or room, if going passive two way REW via soundcard and a cheap resistor can measure impedance of drivers sitting in their box which can help make the passive XO network more precise. Link REW - Room EQ Wizard Room Acoustics Software. Will work on Windows and Mac.
Link to knowledge around baffle step and diffraction choose topic 2 there True Audio's Loudspeaker Design Tech Topics Index.
To simulate baffle step and diffraction link to great free Swedish tool choose The Edge and download Tolvan Data. Will work on Windows.
To simulate baffle step and diffraction plus room modes link to free Excel application software and choose "Baffle Diffraction and Boundary Simulator". Will work on Windows and needs MS Excel license both old and new versions are supported.
If going two way to find best XO point try JRiver mediacenter where its DSP engine have all plugins build in, it's not meant as your future setup just to find the best XO point and slopes for you drivers that then later can be build as passive ones. In the route set up the DSP engine to mix L and R to mono then put a choosed LP XO slope to route to left output and a choosed HP XO slope to route to right output. Now your power amps left channel shall feed full ranger and right channel feed tweeter. On the fly you can change filters slope and frequency to find best ones without buying real components, listen and measure to a mono speaker and you could even add EQ if wanted. Link JRiver Media Center software. Works on Windows and Mac, although unsure if Mac has as many plugins as Windows version, the trail version have full function in a month. Tip to measure with REW and have JR engine filters looped into Windows sound system, choose the virtual sound card "JRiver Media Center 20" as default soundcard and keep JR program open in back ground, then all programs making sound will be looped through what ever settings JR DSP engine add.
Passive XO simulator to tune passive XO, very good to reach precision of a certain intended slope. Link http://www.diyaudio.com/forums/multi-way/259865-xsim-free-crossover-designer.html. Work on Windows probably need some exercise time but think is worth the time.
Calibrated microphone at 89€ a German webshop, USB interface works on Windows and Mac. Link miniDSP USB Messmikrofon UMIK-1 | Lautsprcher Selbstbau.
Hope links not dead or pointing wrong 🙂.
Last edited:
I suggest connecting the supertweeter with an adjustable L-pad. Then dial it up until you can just hear it, and turn it down a notch. Read both legs of the L-pad with a multimeter, and rebuild it with fixed audiophile quality resistors, if desired.
This allows you to test a bunch of different highpass cap values quickly. The method is an old school way of adding a supertweeter, but it has value.
For learning purposes, IME it's good to fiddle with it by ear for a time, then get some measuring gear and find out what's really happening, then do it again.
This allows you to test a bunch of different highpass cap values quickly. The method is an old school way of adding a supertweeter, but it has value.
For learning purposes, IME it's good to fiddle with it by ear for a time, then get some measuring gear and find out what's really happening, then do it again.
- Status
- Not open for further replies.