Hi CLS.
Several subscribers have asked for an OB section but it has not happened, and as the T-bass can work with a single FR driver too I put it here.
We NEED on OB section !
You wrote;-
>> On the other hand, I noticed the impedance of the transformer winding is peaked at about 2kHz, and then drops. How so? Is it because of too much parasitical capacitance? It's quite different from 'normal' chokes with simple exponential rise curves.... <<
Without more details I cannot understand exactly what is happening here.
Are you noting mpedance reflected to the amp or the driver ?
Or is the the transformer being energised in isolation without the parallel connection of the series C+R across one winding and which takes over above 100Hz anyway.
Yes it is 'iron'ic that this circuit brings us back to transformers with SS, but it is due to LF driver requirements.
(Any wonder bass guitarists prefer tube amplifiers - global voltage NFB/SS amps do not (cannot) behave the same around driver Fs.)
Cheers ......... Graham.
Several subscribers have asked for an OB section but it has not happened, and as the T-bass can work with a single FR driver too I put it here.
We NEED on OB section !
You wrote;-
>> On the other hand, I noticed the impedance of the transformer winding is peaked at about 2kHz, and then drops. How so? Is it because of too much parasitical capacitance? It's quite different from 'normal' chokes with simple exponential rise curves.... <<
Without more details I cannot understand exactly what is happening here.
Are you noting mpedance reflected to the amp or the driver ?
Or is the the transformer being energised in isolation without the parallel connection of the series C+R across one winding and which takes over above 100Hz anyway.
Yes it is 'iron'ic that this circuit brings us back to transformers with SS, but it is due to LF driver requirements.
(Any wonder bass guitarists prefer tube amplifiers - global voltage NFB/SS amps do not (cannot) behave the same around driver Fs.)
Cheers ......... Graham.
Graham Maynard said:
Yes, that impedance behavior came from the transformer alone, without any other components or circuits. I found it when I was measuring the windings for their inductance and impedance.
I saw such behavior on several cases.
As to the SS amp, I didn't mention one thing before. My good old Hafler was heavily modified. One of the mods was the global NFB circuit -- I changed the feedback point to the prior driver stage (instead of the 'normal' output point). So it's no longer a "normal" SS amp with GNFB. The reason for this is decoupling the back EMF current from the driver, so it'd not mess up with the whole circuit round and round...
I like the sound after such mod, so I keep it that way. Now with T-bass, I don't notice any ill effect becuase of it, luckily.
Haha, it seems I'm a follower and also a trouble maker
I re-routed the feedback of the amp to the 'normal' output point last night. And it's obvious the mid to lower bass got stronger (more output). I don't have equipment for precise measurement for this. My gut feeling is it gains 2~4dB across the bass range (maybe even more, I don't know). I guess this would be the effect of 'tightly feedback from the load' -- drive the load regardless of load change.
Sound quality wise, there's no surprise. By only a brief listening, I felt it just delivered more output, with almost equal solidity and authority. In fact, the bass was too much, and I ran out of adjustment range on the upstream EQ. So the overall balance was too bass-heavy, thus hard to verify the true 'sound quality'.
Guess it'd be a good thing, though -- more bass output with all others being equal. It kind of means more positive effect brought by the T-bass circuit.
Now I'm going to change the amp for mid-high to get more gain and balance the whole system...
By the way, this is the picture of the new 400VA toroids:
The size is 11.5cm in diameter, total "800VA" capacity in there. Meantime, I think the power transformer in my little old Hafler is just 300VA or so....
Sound quality wise, there's no surprise. By only a brief listening, I felt it just delivered more output, with almost equal solidity and authority. In fact, the bass was too much, and I ran out of adjustment range on the upstream EQ. So the overall balance was too bass-heavy, thus hard to verify the true 'sound quality'.
Guess it'd be a good thing, though -- more bass output with all others being equal. It kind of means more positive effect brought by the T-bass circuit.
Now I'm going to change the amp for mid-high to get more gain and balance the whole system...
By the way, this is the picture of the new 400VA toroids:
The size is 11.5cm in diameter, total "800VA" capacity in there. Meantime, I think the power transformer in my little old Hafler is just 300VA or so....
Nice breadboarding there CLS.
I don't see any resistors with the toroidal version though; this can lead to too much or overpowering bass.
Resistors are essential to tame the first half cycle boost which can make drum and bass guitar picking seem overemphasised and eventually tiring.
Do try one ohm starting values in series with each of the capacitor and choke and listen to the way only those who actually construct the circuit can hear that first half cycle compensation actually arising.
( An effect which simply cannot be revealed via textbook sinewave frequency response sweeps nor be evaluated by conventional software !!! )
Yes the 400VA components are large, but you can get output to below 25Hz on OB with them, and with 18" drivers maybe even a respectable 20Hz.
Cheers ........ Graham.
I don't see any resistors with the toroidal version though; this can lead to too much or overpowering bass.
Resistors are essential to tame the first half cycle boost which can make drum and bass guitar picking seem overemphasised and eventually tiring.
Do try one ohm starting values in series with each of the capacitor and choke and listen to the way only those who actually construct the circuit can hear that first half cycle compensation actually arising.
( An effect which simply cannot be revealed via textbook sinewave frequency response sweeps nor be evaluated by conventional software !!! )
Yes the 400VA components are large, but you can get output to below 25Hz on OB with them, and with 18" drivers maybe even a respectable 20Hz.
Cheers ........ Graham.
Thanks Graham
Yes, the 'no series resistor' version above (with only 470uF) was indeed a little bit over -- by ears and by impedance scan. It's impedance showed the lowest point of 1 Ohm at 100Hz (28 Ohm peak @ 18.8Hz). It supposedly sucked the most current at around that lowest point and produced chest punding midbass. RTA proved this. I had to add some more attenuations on EQ to calm it down, so it's a waste.
Later, by some simple calcuations and comparisons of the reactances of C & L, I tried bigger C (1800uF) with 1 Ohm resistor in series (still no R with choke). This broadened and flattened the boost and extend lower. The impedance scan showed the lowest point moved to 50Hz (same low 1 Ohm!), with a much tamed peak of only 10 Ohm @ teens!
(The woofer itself on baffle shows 170 Ohm @ 19.5Hz, after T-bass'ed, it's 5 Ohm at the same point. What a difference here! )
That was obvious audible and also proved by RTA -- a good move of making the boost flatter and lower.
On the other hand, I dug out other amp for mid-high to replace the previous beloved 0.7w single stage 6S45 spud amp which was not able to deliver sufficient gain. For leveling the fullrange balance, I attenuated A LOT in the bass range by the upstream digital EQ. As mentioned, it's a waste of gain and also some loss in resolution.
With more gain on mid-high, the adjustments on EQ were dramatically reduced. This helped a lot in alomst every aspect. The mid-high got more energy and livelier and the bass can also give its best shot without holding back -- a big and overall improvement.
Jumpy & lively bass has been mentioned too many times. I heard 'new' things in recordings! Like a child with a new toy, I dug out more and more CD and DVD, played them eagerly. Those recordings I thought with muddy bass are now SO MUCH better. And those already good are now even better, I'm thrilled. (but of course there're still some bad recordings proved to be cureless).
I must say this is one of the biggest upgrade in my whole audio life (20 years now).
You're the man! Thank you so much.
Yes, the 'no series resistor' version above (with only 470uF) was indeed a little bit over -- by ears and by impedance scan. It's impedance showed the lowest point of 1 Ohm at 100Hz (28 Ohm peak @ 18.8Hz). It supposedly sucked the most current at around that lowest point and produced chest punding midbass. RTA proved this. I had to add some more attenuations on EQ to calm it down, so it's a waste.
Later, by some simple calcuations and comparisons of the reactances of C & L, I tried bigger C (1800uF) with 1 Ohm resistor in series (still no R with choke). This broadened and flattened the boost and extend lower. The impedance scan showed the lowest point moved to 50Hz (same low 1 Ohm!), with a much tamed peak of only 10 Ohm @ teens!
(The woofer itself on baffle shows 170 Ohm @ 19.5Hz, after T-bass'ed, it's 5 Ohm at the same point. What a difference here! )
That was obvious audible and also proved by RTA -- a good move of making the boost flatter and lower.
On the other hand, I dug out other amp for mid-high to replace the previous beloved 0.7w single stage 6S45 spud amp which was not able to deliver sufficient gain. For leveling the fullrange balance, I attenuated A LOT in the bass range by the upstream digital EQ. As mentioned, it's a waste of gain and also some loss in resolution.
With more gain on mid-high, the adjustments on EQ were dramatically reduced. This helped a lot in alomst every aspect. The mid-high got more energy and livelier and the bass can also give its best shot without holding back -- a big and overall improvement.
Jumpy & lively bass has been mentioned too many times. I heard 'new' things in recordings! Like a child with a new toy, I dug out more and more CD and DVD, played them eagerly. Those recordings I thought with muddy bass are now SO MUCH better. And those already good are now even better, I'm thrilled. (but of course there're still some bad recordings proved to be cureless).
I must say this is one of the biggest upgrade in my whole audio life (20 years now).
You're the man! Thank you so much.
Hi Simon,
Well, the next project will be panomaniac's Manzanita. At the moment I have problem to get funds for this.
To put it mildly, it is unfortunately as I will be home the whole December
I was merely curious if it is doable! I have four Vifa P21 WO 12-08 collecting dust. When all other things been fixed, I will get my hands on some transformers for sure.
Peter
Well, the next project will be panomaniac's Manzanita. At the moment I have problem to get funds for this.
To put it mildly, it is unfortunately as I will be home the whole December
I was merely curious if it is doable! I have four Vifa P21 WO 12-08 collecting dust. When all other things been fixed, I will get my hands on some transformers for sure.
Peter
Hi CLS,
For me its the best bass after 40 years when I had been doing Disco with 4x 18"ers (Plus 8s and tweeters) powered by tube amps.
How can any enclosured loudspeaker get close to this ?
Our LF drivers can be displacing with considerable excursion and yet they don't sound muddy.
One problem here being that most folk do not appreciate just how muddy the bass is that they are already enjoying.
Where before the weaknesses of OB were deemed to be driver related, this circuit shows that it is really down to the direct connection between driver and SS amplifier, or a conventional crossover network increasing series impedance below FS and further muddying reproduction due to the capacitor across the driver introducing phase shift and interacting with driver stiored energies.
Tube amps can sound better and maybe louder, but they cannot maintain the same control of LF definiyion.
(Keep tubes for 100Hz upwards.)
The impedance dip you observe is down to the series L and C resonant circuit behaving like a tuned shunt across the amplifier output terminals. This might be somewhat avoidable but I'm not able to investigate further, and besides it does not worry me because any decent amplifier should cope anyway.
It is the way that the tuned circuit interacts with the driver which counts.
Hi Peter,
Yes the T-bass will work with boxes, however you can get sufficient and better quality bass for home use from a large driver in say a U-frame of similar volume to a floorstanding boxed LS.
Once you hear how clean the deep bass can become you'll not want to use boxes again.
For me not even Subs either, this because their room pressurisation only makes things worse.
Try two of your P21s in a backless box (U-frame) as large as you can accept, with T-bass.
Then you can try fitting a back, but I know you won't want to after you've heard it !!!
Cheers ........Graham.
For me its the best bass after 40 years when I had been doing Disco with 4x 18"ers (Plus 8s and tweeters) powered by tube amps.
How can any enclosured loudspeaker get close to this ?
Our LF drivers can be displacing with considerable excursion and yet they don't sound muddy.
One problem here being that most folk do not appreciate just how muddy the bass is that they are already enjoying.
Where before the weaknesses of OB were deemed to be driver related, this circuit shows that it is really down to the direct connection between driver and SS amplifier, or a conventional crossover network increasing series impedance below FS and further muddying reproduction due to the capacitor across the driver introducing phase shift and interacting with driver stiored energies.
Tube amps can sound better and maybe louder, but they cannot maintain the same control of LF definiyion.
(Keep tubes for 100Hz upwards.)
The impedance dip you observe is down to the series L and C resonant circuit behaving like a tuned shunt across the amplifier output terminals. This might be somewhat avoidable but I'm not able to investigate further, and besides it does not worry me because any decent amplifier should cope anyway.
It is the way that the tuned circuit interacts with the driver which counts.
Hi Peter,
Yes the T-bass will work with boxes, however you can get sufficient and better quality bass for home use from a large driver in say a U-frame of similar volume to a floorstanding boxed LS.
Once you hear how clean the deep bass can become you'll not want to use boxes again.
For me not even Subs either, this because their room pressurisation only makes things worse.
Try two of your P21s in a backless box (U-frame) as large as you can accept, with T-bass.
Then you can try fitting a back, but I know you won't want to after you've heard it !!!
Cheers ........Graham.
Hi Graham,
I have tried, sort of, with ob subwoofer. By a mysterius reason I do not understand today, I bought a Sunfire with the fabolous name "True subwoofer MKII" couple of 10 years ago.
I removed the slave and let it play open in a corner. Heaps better than the original "mush"
I will try with mine Vifa woffers when all is done and over!! Right now I have 130L boxes in the room so no problem with size My wife like huge thingies
Peter
I have tried, sort of, with ob subwoofer. By a mysterius reason I do not understand today, I bought a Sunfire with the fabolous name "True subwoofer MKII" couple of 10 years ago.
I removed the slave and let it play open in a corner. Heaps better than the original "mush"
I will try with mine Vifa woffers when all is done and over!! Right now I have 130L boxes in the room so no problem with size
My wife like huge thingies Peter
Hi Graham,
First of all: I believe that your circuit is a great and innovative idea and I definitely intend to try it soon! However, i would like to make some comments to your discussion with sreten at the beginning of the thread:
- I agree with you that the analysis of only the electrical behavior of the circuit on a dummy resistive load(or even a more complex electrical model) is not able to tell the whole story - for me, circuit+driver needs to be analyzed, since they are tightly coupled.
- However, sreten's simulation is useful in having an approximate picture of the load seen by the amp.
- Your concept of "first half-cycle" is a useful and intuitive way to explain things, but I do not think that it is really necessary: I am not an expert in this field, but AFAIK the transient response of a system (because that's what the first cycle is!) is a well studied thing: impulse or step response should pretty much contain everything one needs to know! But again, the whole system circuit+woofer(ACOUSTIC output, since that is what matters in the end) needs to be analyzed and I believe the easiest way to do it would be by means of measuring the impulse response and maybe comparing to the trivial way of EQ-ing the woofer at line level ?
- as to emulating your circuit at line-level: I also believe that this would be very difficult to achieve (although theoretically not impossible). As we all know, a drawback of a passive crossover over an active one is that its electrical corrections are strongly dependent of the complex load a loudspeaker driver represents: in this case, though, you seem to have turned this to work FOR you and compensate the inherent driver misbehavior at and below resonance.
I know you actually suggest building the circuit first and then talking: however I truly believe that some more analysis/measurements should help clarifying things further.
First of all: I believe that your circuit is a great and innovative idea and I definitely intend to try it soon! However, i would like to make some comments to your discussion with sreten at the beginning of the thread:
- I agree with you that the analysis of only the electrical behavior of the circuit on a dummy resistive load(or even a more complex electrical model) is not able to tell the whole story - for me, circuit+driver needs to be analyzed, since they are tightly coupled.
- However, sreten's simulation is useful in having an approximate picture of the load seen by the amp.
- Your concept of "first half-cycle" is a useful and intuitive way to explain things, but I do not think that it is really necessary: I am not an expert in this field, but AFAIK the transient response of a system (because that's what the first cycle is!) is a well studied thing: impulse or step response should pretty much contain everything one needs to know! But again, the whole system circuit+woofer(ACOUSTIC output, since that is what matters in the end) needs to be analyzed and I believe the easiest way to do it would be by means of measuring the impulse response and maybe comparing to the trivial way of EQ-ing the woofer at line level ?
- as to emulating your circuit at line-level: I also believe that this would be very difficult to achieve (although theoretically not impossible). As we all know, a drawback of a passive crossover over an active one is that its electrical corrections are strongly dependent of the complex load a loudspeaker driver represents: in this case, though, you seem to have turned this to work FOR you and compensate the inherent driver misbehavior at and below resonance.
I know you actually suggest building the circuit first and then talking: however I truly believe that some more analysis/measurements should help clarifying things further.
Hi,
I'd also like to make some more actual measurements on the circuit if I can. Now I can only make some simple impedance scans but can not see the actual current flow through individual paths...
I'm no expert in analysing circuits. But I'm a bass freak. In the "heyday", I got 6 x 18" in my little living room. I believe very few or simply no sane people do this. Last time I posted the pictures of my bass baffle and got this question: "are you married?" (yes)
With all those cone areas, I didn't need very much amp power or even cabinet to get a lot of bass. By some helps from EQ, I can get very extended response in room, too (again, no sweat on those big woofers). I got pretty much output all the way down to 20~25Hz range, OB. Measurements were indeed rough, but RTA did show the output was there, and my hearing and vibrations on body (and windows etc) also told me that deep bass really exist.
But those were not really satisfying, TBH.
Now I'm so happy with a pair of mid size OB made of lame plywood and only 1 woofer per side, AND T-BASS! It really rocks!
Build it, please! Nothing to loose. All parts can be recycled if regret. And you won't.
I'd also like to make some more actual measurements on the circuit if I can. Now I can only make some simple impedance scans but can not see the actual current flow through individual paths...
I'm no expert in analysing circuits. But I'm a bass freak. In the "heyday", I got 6 x 18" in my little living room. I believe very few or simply no sane people do this. Last time I posted the pictures of my bass baffle and got this question: "are you married?" (yes)
With all those cone areas, I didn't need very much amp power or even cabinet to get a lot of bass. By some helps from EQ, I can get very extended response in room, too (again, no sweat on those big woofers). I got pretty much output all the way down to 20~25Hz range, OB. Measurements were indeed rough, but RTA did show the output was there, and my hearing and vibrations on body (and windows etc) also told me that deep bass really exist.
But those were not really satisfying, TBH.
Now I'm so happy with a pair of mid size OB made of lame plywood and only 1 woofer per side, AND T-BASS! It really rocks!
Build it, please! Nothing to loose. All parts can be recycled if regret. And you won't.
Hi CLS,
I 've seen some of your walls, sorry, I mean OB's
How big is medium for you? More over, what about the difference with T-bass versus eq.
I have maybe a stupid question. Most T-bass circuits I've seen are 110V's. Will it make a difference with 240's?-impedance of coils?
I have been searching for toroids on internet but have not found specs regarding impedance of the windings. Lundahl was the only specified so far but jeez, they are pricey... Anyone have suggestions of web stores in Europe? Meanwhile I will have a look here in Brno tomorrow... If I find suitable toroids, it will not be so expensive as buying new drivers also. They will come later when Mammon is back.
Peter
I 've seen some of your walls, sorry, I mean OB's
How big is medium for you? More over, what about the difference with T-bass versus eq.
I have maybe a stupid question. Most T-bass circuits I've seen are 110V's. Will it make a difference with 240's?-impedance of coils?
I have been searching for toroids on internet but have not found specs regarding impedance of the windings. Lundahl was the only specified so far but jeez, they are pricey... Anyone have suggestions of web stores in Europe? Meanwhile I will have a look here in Brno tomorrow... If I find suitable toroids, it will not be so expensive as buying new drivers also. They will come later when Mammon is back.
Peter
Hi sreten,
Doh, the first post showing the circuit got me confused
http://www.diyaudio.com/forums/attachment.php?s=&postid=1622325&stamp=1222862646
Thanks,
Peter
Doh, the first post showing the circuit got me confused
http://www.diyaudio.com/forums/attachment.php?s=&postid=1622325&stamp=1222862646
Thanks,
Peter
Hi Peter,
The OB I'm using now is measured 120cm H x 90cm W, with 1 x 18" woofer, 1 x 8" mid , and 1 x 8" WG with 1" driver. I believe it can be narrower and keep most of the performance (at least in my little living room, I guess).
For the previous EQ'ed OB bass, I used shallow 1st order active xover between low and mid at about 160Hz, plus some EQ to make it reasonably flat. EQ for bass has 2 parts: one is 6dB/oct low pass start from 40Hz (for compensating the baffle loss), the other is 'fine tuning' on digital EQ, adjustments were mostly in the range of -6dB ~ +3dB or so.
It played very low and loud without problem, but it was just not very coherent with the mid-high. Sound from the bass alone was clear enough, but gentle, even light as a breeze, not very heavy-footed as the size suggested. Bass notes stop quickly enough, as it should be for a low Q woofer without box, but I got a feeling that the "startup" was not quick enough. So the sound lacks the sense of punch or jump.
Actually I was quite disappointed when I got it built. I've tried to make it tuneful as much as I can, with various methods of either an engineer or an 'audiophile'. Eventually I thought to myself, maybe this is it. Well, I'm done.
Until I tried T-bass.
I heard the difference right at the first mockup version of T-bass, and that was way way way out of the optimum. It is the sense of convergency of timing -- the timing when mid and low start. Once they are focused, the overall sound change a lot. Mid to midbass to low bass sounds are often combined together as multiple overtones etc. When they play coherently, the sounds come out as micro explosions.
Listening to T-bass these days, it keeps reminding me of horn bass. They have similar intense and solid punches. Very fast rise and fall, start and stop. I think a bass horn gets the gain accoustically while T-bass obtains similar result by sucking extra current from amp.
I love bass horn a lot, just can not use it in home. Now I'm not sorry for that any more.
The OB I'm using now is measured 120cm H x 90cm W, with 1 x 18" woofer, 1 x 8" mid , and 1 x 8" WG with 1" driver. I believe it can be narrower and keep most of the performance (at least in my little living room, I guess).
For the previous EQ'ed OB bass, I used shallow 1st order active xover between low and mid at about 160Hz, plus some EQ to make it reasonably flat. EQ for bass has 2 parts: one is 6dB/oct low pass start from 40Hz (for compensating the baffle loss), the other is 'fine tuning' on digital EQ, adjustments were mostly in the range of -6dB ~ +3dB or so.
It played very low and loud without problem, but it was just not very coherent with the mid-high. Sound from the bass alone was clear enough, but gentle, even light as a breeze, not very heavy-footed as the size suggested. Bass notes stop quickly enough, as it should be for a low Q woofer without box, but I got a feeling that the "startup" was not quick enough. So the sound lacks the sense of punch or jump.
Actually I was quite disappointed when I got it built. I've tried to make it tuneful as much as I can, with various methods of either an engineer or an 'audiophile'. Eventually I thought to myself, maybe this is it. Well, I'm done.
Until I tried T-bass.
I heard the difference right at the first mockup version of T-bass, and that was way way way out of the optimum. It is the sense of convergency of timing -- the timing when mid and low start. Once they are focused, the overall sound change a lot. Mid to midbass to low bass sounds are often combined together as multiple overtones etc. When they play coherently, the sounds come out as micro explosions.
Listening to T-bass these days, it keeps reminding me of horn bass. They have similar intense and solid punches. Very fast rise and fall, start and stop. I think a bass horn gets the gain accoustically while T-bass obtains similar result by sucking extra current from amp.
I love bass horn a lot, just can not use it in home. Now I'm not sorry for that any more.
Although I am going to try the circuit myself, I'll play a little bit on the devil's advocate part on this one - a healthy dose of skepicism.
Sometimes - I am not saying it's the case here - the "spectacular improvements" people claim are psychoacoustic effects that are perceived in a certain way and actually originate in acoustic anomalies in amplitude or plain distortion.
Yes, the explanation of the way T-Bass works seems to indicate that the circuit has certain clear advantages over plain dumb EQ. However some measurement and testing would be necesary to confirm these advantages.
One psychoacoustic phenomen that I vaguely remember reading about in this forum was that it takes many wave cycles for our hearing to actually perceive low frequencies: this would somewhat be against trying to improve the "first half cycle", as this simply would not matter. However, it would NOT disable the whole concept of improving the transient response and group delay.
Again please do not misunderstand me: I truly believe the circuit has potential and I will give it a try soon (only a bit worried about handlig the very low impedance, not sure any of my so-so amps would handle it)
Sometimes - I am not saying it's the case here - the "spectacular improvements" people claim are psychoacoustic effects that are perceived in a certain way and actually originate in acoustic anomalies in amplitude or plain distortion.
Yes, the explanation of the way T-Bass works seems to indicate that the circuit has certain clear advantages over plain dumb EQ. However some measurement and testing would be necesary to confirm these advantages.
One psychoacoustic phenomen that I vaguely remember reading about in this forum was that it takes many wave cycles for our hearing to actually perceive low frequencies: this would somewhat be against trying to improve the "first half cycle", as this simply would not matter. However, it would NOT disable the whole concept of improving the transient response and group delay.
Again please do not misunderstand me: I truly believe the circuit has potential and I will give it a try soon (only a bit worried about handlig the very low impedance, not sure any of my so-so amps would handle it)