I did both. They are not identical but very close. And the numbers are repeatable by the WT3 (again, not precisely identical in each measurement, but very close). Good enough for me.
Sorry, got the impression that there was nothing wrong with your meter.
Those cheap LCR-meters out there only measures the impedance of the inductor with a very small signal, generated by a too high impedance source.
I've been browsing through this thread since CLS recommended I should use the T-bass circuit on my dipoles. Its been hard to figure out what it actually does, apart from some EQ. The thing that I dont understand is how it can affect the first half of the wave period (as claimed) and do dynamic correction.
It would however be quite simple to measure if it works or not. Anyone tried to do this? An approach would be to do the same thing Dynaudio did to show off the dynamic behaviour of their tweeters: Record the driver output of three periods of sine wave, and compare that to the source signal. That should tell it all, or not?
As usual, take this post as a rambling, un-scientifically opinion....
I have assembled the circuit and tested on some "nude speakers".
Initially it was a disappointment, couldn't notice any big difference. I used these values at first...
The 'T'-bass
I decided then to try the circuit on my Sonido boxes. I didn't dare to connect to my Naim amp, so I took my "stand-in" NAD 7020 reciever.
After playing with the caps and resistance for a while, I settled on 1500uF and 0.5 ohm When using 1 ohm, the effect was very subtle. The Sonido boxes are peaking 6-8db at ~150Hz and T-bass with above values, evened out the response a bit.
However, when playing sinus wave at 20Hz, it is visible that with T-bass, the stroke of cone is halved?? (Could be good for widerange drivers)
I remember an earlier post that described same phenomen. The output increases a little bit at 30Hz, while the output between 40-70Hz is definitely increased.
There is more punch now but I have a hard time to define if this circuit actually "re-create" the initial attack of bass or drums, or simply just boost bass as an equalizer.
As I understand, my amp "see's" a ~0.5 ohm load now. It does not seem to bother the NAD amp. I was playing music the whole Sunday and it didn't get any warmer than usual.
More to come, if I have time next weekend.
Peter
I have assembled the circuit and tested on some "nude speakers".
Initially it was a disappointment, couldn't notice any big difference. I used these values at first...
The 'T'-bass
I decided then to try the circuit on my Sonido boxes. I didn't dare to connect to my Naim amp, so I took my "stand-in" NAD 7020 reciever.
After playing with the caps and resistance for a while, I settled on 1500uF and 0.5 ohm
When using 1 ohm, the effect was very subtle. The Sonido boxes are peaking 6-8db at ~150Hz and T-bass with above values, evened out the response a bit. However, when playing sinus wave at 20Hz, it is visible that with T-bass, the stroke of cone is halved?? (Could be good for widerange drivers)
I remember an earlier post that described same phenomen. The output increases a little bit at 30Hz, while the output between 40-70Hz is definitely increased.
There is more punch now but I have a hard time to define if this circuit actually "re-create" the initial attack of bass or drums, or simply just boost bass as an equalizer.
As I understand, my amp "see's" a ~0.5 ohm load now. It does not seem to bother the NAD amp. I was playing music the whole Sunday and it didn't get any warmer than usual.
More to come, if I have time next weekend.
Peter
Hi Peter,
So you are using Sonido boxes. Resistance/ Impedance at LF ?
The 5mH section was to assist with LF driver crossover on a baffle; you using a low R coil ?
Try leaving this section out until you understand everything happening in the 'T' part.
I trust your transformer has low resistance windings.
How can your amp be 'Seeing' ~0.5 ohm ?
>> I have a hard time to define if this circuit actually "re-create" the initial attack of bass or drums, or simply just boost bass as an equalizer. <<
You cannot have LF 'boost' without phase change from turnover and on.
This circuit uses a transformer to step-up LF voltage *without LF phase change* at the boosted frequencies. When you realise you CAN hear the difference between both scenarios, you will realise who it really is who rambles unscientifically when they claim we are incapable of discerning reproduction phase at low bass frequencies.
Don't be expecting more than this passive circuit is capable of acheiving, especially with an enclosure loaded driver !
A low Z-out amplifier and thick cables are essential too.
Cheers ......... Graham.
So you are using Sonido boxes. Resistance/ Impedance at LF ?
The 5mH section was to assist with LF driver crossover on a baffle; you using a low R coil ?
Try leaving this section out until you understand everything happening in the 'T' part.
I trust your transformer has low resistance windings.
How can your amp be 'Seeing' ~0.5 ohm ?
>> I have a hard time to define if this circuit actually "re-create" the initial attack of bass or drums, or simply just boost bass as an equalizer. <<
You cannot have LF 'boost' without phase change from turnover and on.
This circuit uses a transformer to step-up LF voltage *without LF phase change* at the boosted frequencies. When you realise you CAN hear the difference between both scenarios, you will realise who it really is who rambles unscientifically when they claim we are incapable of discerning reproduction phase at low bass frequencies.
Don't be expecting more than this passive circuit is capable of acheiving, especially with an enclosure loaded driver !
A low Z-out amplifier and thick cables are essential too.
Cheers ......... Graham.
Last edited:
You can have LF boost/EQ without phase shift, its called digital Phase-Linear EQ.
If the T-bass circuit worked as a pure transformer, then yes - there would be no phase shift. Problem is that it *does* change the frequency response, and that also means it will introduce phase shift. In the analog world, any circuit that changes the frequency response also changes the phase.
It might be that active analog or digital phase-warping EQ changes the phase a lot more than the T-bass circuit, which indeed is a valid argument for using the T-bass circuit.
If the T-bass circuit worked as a pure transformer, then yes - there would be no phase shift. Problem is that it *does* change the frequency response, and that also means it will introduce phase shift. In the analog world, any circuit that changes the frequency response also changes the phase.
It might be that active analog or digital phase-warping EQ changes the phase a lot more than the T-bass circuit, which indeed is a valid argument for using the T-bass circuit.
Hi StigErik,
Do you have any starting from zero Input-Output sine waveform traces for the digital phase linear EQ at say 15Hz, 30Hz, 60Hz 120Hz for LF lift coming in with 50Hz turnover ?
Amplitude, phase, waveform in real time ?
I said there was no T-bass phase change in the boosted region. Besides the real advantage of the T-bass is not phase or boost related; it is the inverse nature of the tuning and loading via the centre limb which modifies transformer action, such that cut is lossily induced by a reactance adjusted to act inversely to the driver system reactance - not like a phase changing passive crossover !
No ordinary EQ can do that; active, passive or digital !
Do you have any starting from zero Input-Output sine waveform traces for the digital phase linear EQ at say 15Hz, 30Hz, 60Hz 120Hz for LF lift coming in with 50Hz turnover ?
Amplitude, phase, waveform in real time ?
I said there was no T-bass phase change in the boosted region. Besides the real advantage of the T-bass is not phase or boost related; it is the inverse nature of the tuning and loading via the centre limb which modifies transformer action, such that cut is lossily induced by a reactance adjusted to act inversely to the driver system reactance - not like a phase changing passive crossover !
No ordinary EQ can do that; active, passive or digital !
Last edited:
I dont. But - it would be very interesting to check it out, and do the same with the T-bass circuit.
Yes, in a minimum-phase system, the phase is flat when the frequency response is flat. Simple as that.
I still dont understand that part. Maybe because a simple, trustworthy and good explanation has not been presented yet.. one that a layman like me can understand.
Doing waveform measurements will shed some light on this I think.
There is nothing worse than a personally aimed challenge by someone who does not have hands-on *real world* experience of what they are writing about.
It is clear you do not understand the sheer simplicity of what this circuit does, and your failure to understand it is not my fault, I have explanation somewhere !
If you cannot understand some part of the explanation then why not ask, instead of saying that what I have written is not simple, not good or not *trustworthy*.
Please go and simulate this circuit before you make more meaningless challenges like
>> Yes, in a minimum-phase system, the phase is flat when the frequency response is flat. Simple as that. <<
FACT: The amplitude at the driver *DOES* change with frequency when the phase does not.
Go figure ! It really is very simple !
Yes doing waveform tests will shed light - I already have.
The circuit is given freely for anyone to try, so why don't you before you cast more offensive dispersions.
.
It is clear you do not understand the sheer simplicity of what this circuit does, and your failure to understand it is not my fault, I have explanation somewhere !
If you cannot understand some part of the explanation then why not ask, instead of saying that what I have written is not simple, not good or not *trustworthy*.
Please go and simulate this circuit before you make more meaningless challenges like
>> Yes, in a minimum-phase system, the phase is flat when the frequency response is flat. Simple as that. <<
FACT: The amplitude at the driver *DOES* change with frequency when the phase does not.
Go figure ! It really is very simple !
Yes doing waveform tests will shed light - I already have.
The circuit is given freely for anyone to try, so why don't you before you cast more offensive dispersions.
.
Graham - I am sorry that you feel I am stupid, ignorant, provide meaningless input and so on .... but can you PLEASE point me to the waveform tests you mention? Thanks. Thats all I ask. I'm trying to learn here, so if you want to be a good teacher - please put up with your nagging students and get the message through.....
Hi Graham,
Sorry, nothing else than nom resistance. I will check if I have more on specs papers.
I don't use it with the Sonido's, it's there for later with ob bass speakers.
Coils are around 0.1 ohm.
I measure ~0.5-0.6 ohm over terminals?! It is hard to say, as impedance meter is not accurate at that low resistances. Maybe wrongly connected, but my thick head can't find anything wrong with the circuit...
Right now I'm in Germany so no further checking from me.
Cheers,
Peter
Sorry, nothing else than nom resistance. I will check if I have more on specs papers.
I don't use it with the Sonido's, it's there for later with ob bass speakers.
Coils are around 0.1 ohm.
I measure ~0.5-0.6 ohm over terminals?! It is hard to say, as impedance meter is not accurate at that low resistances. Maybe wrongly connected, but my thick head can't find anything wrong with the circuit...
Right now I'm in Germany so no further checking from me.
Cheers,
Peter
Hi StigErik,
I have so many other problems to cope with that I have no wish to be a teacher.
I no longer remember where any waveforms are posted either. If not here, then maybe Audio Circle. I don't even know where they might be here on computer Roms due to changing computer operating systems.
I follow this thread to help constructors out, not armchair challengers.
A loudspeaker cone does not transduce amplitude linearly in time from a suddenly starting note (bass drum/guitar). (See Linkwitz.) It stores energy during the first 90 waveform degrees and releases it later in LOUDSPEAKER time, not music signal time.
The T-bass can be tuned to do the opposite in not reducing output during the first 90 degrees (actually transformer boosts it), but presenting impedance and thus coincidentally reducing drive as the cone becomes resonant.
When one circuit (T-bass electrical) operates as a close inverse to the other (driver electrodynamic), a time variable driver amplitude compensation can arise during music time with minimum phase change.
The 'boost' is compensation for reactive characteristics during music signal time, not some constant external amplitude modification unavoidably having constant effect upon all LF waveforms.
Also the T-bass can be effective with typical OB drivers due to them having greater impedance below resonance than do typically boxed drivers (often damped). LF voltage is stepped up to counter increased LF OB impedance such that one full range amp can still drive an entire baffle.
.
I have so many other problems to cope with that I have no wish to be a teacher.
I no longer remember where any waveforms are posted either. If not here, then maybe Audio Circle. I don't even know where they might be here on computer Roms due to changing computer operating systems.
I follow this thread to help constructors out, not armchair challengers.
A loudspeaker cone does not transduce amplitude linearly in time from a suddenly starting note (bass drum/guitar). (See Linkwitz.) It stores energy during the first 90 waveform degrees and releases it later in LOUDSPEAKER time, not music signal time.
The T-bass can be tuned to do the opposite in not reducing output during the first 90 degrees (actually transformer boosts it), but presenting impedance and thus coincidentally reducing drive as the cone becomes resonant.
When one circuit (T-bass electrical) operates as a close inverse to the other (driver electrodynamic), a time variable driver amplitude compensation can arise during music time with minimum phase change.
The 'boost' is compensation for reactive characteristics during music signal time, not some constant external amplitude modification unavoidably having constant effect upon all LF waveforms.
Also the T-bass can be effective with typical OB drivers due to them having greater impedance below resonance than do typically boxed drivers (often damped). LF voltage is stepped up to counter increased LF OB impedance such that one full range amp can still drive an entire baffle.
.
Hi Peter,
>> I measure ~0.5-0.6 ohm over terminals?! <<
That resistance is just fine, it is the driven impedance which counts.
I covered that aspect in my reply above.
Boxed loudspeakers are damped compared to those for OB working, so whilst the T-bass can still improve LF reproduction via enclosured drivers, the impedances below system resonance are not matched.
Wonder if the Sonidos would reproduce better with their backs removed ?
>> I measure ~0.5-0.6 ohm over terminals?! <<
That resistance is just fine, it is the driven impedance which counts.
I covered that aspect in my reply above.
Boxed loudspeakers are damped compared to those for OB working, so whilst the T-bass can still improve LF reproduction via enclosured drivers, the impedances below system resonance are not matched.
Wonder if the Sonidos would reproduce better with their backs removed ?
This was the impedance curve measured by WT3 when my center OB sub worked with T-bass: http://www.diyaudio.com/forums/multi-way/138614-multiple-dipole-subs-2.html#post1752892
In some frequency range, it did drop below 1 Ohm, the lowest point seemed 0.5 Ohm. The plate amp driving it worked warm, but no failure.
In some frequency range, it did drop below 1 Ohm, the lowest point seemed 0.5 Ohm. The plate amp driving it worked warm, but no failure.