As far as I understand Don, he does not doubt that it works. All he is after is a reliable calculation and/or simulation method before building it. Trial-&-Error is not much in fashion anymore, because it can become very time consuming and expensive, if you don't land the lucky strike at first try. It can also be discouraging, if the first test turns out bad when the component values don't fit.
Hi Pacificblue,
Anyone can simulate this circuit using software, and at least get in the ball-park as far as component values go, however, what a screen shows in technical isolation, CANNOT substitute for human perception requirements related to MUSIC waveforms when reproduced via real-world equipment; no matter how much the 'experts' might think this should be the case !
Anyone can simulate this circuit using software, and at least get in the ball-park as far as component values go, however, what a screen shows in technical isolation, CANNOT substitute for human perception requirements related to MUSIC waveforms when reproduced via real-world equipment; no matter how much the 'experts' might think this should be the case !
Indeed, we all know how unreliable perception is.
But anyway, it's not nice to accuse someone of saying 'it doesn't work' or even accuse him of attacking you personally, if the only thing he tries to get is a better insight.
jd
I have no doubt that I will hear a difference with the T-bass in circuit. And I have little doubt that, provided I have the component values more or less correct, I may well prefer the sound with the T-bass to that without. What I am not prepared to accept (until proven by measurement) is that the reproduced waveform will be more accurate (true to the original) than the reproduction without T-bass. I apply exactly the same argument to, for example, preferences for the sound of a tube amplifier versus a solid-state amplifier.
Two questions which I intend to explain on paper before I burn solder:
- The L-C network is driven from time zero, same as the speaker. Its effect on the speaker drive will start occurring at time zero too, not some time later.
- The additional energy introduced during the first half cycle has to be dissipated in the same time frame (resistively), otherwise it will dissipate later, manifesting itself in increased output of the driver as it does so. StigEric documented this effect using a real speaker and T-bass circuit earlier this month in this forum.
Indeed it did. And, proving my point, it occurred at the time predicted by the mathematical model.
This thread (and the others) has many instances of people trying T-bass using whatever components they have at hand, and as often as not they are less than totally satisfied and begin an expensive process of substituting components. I well remember what the speaker design process was like before Thiele and Small. The T-bass design process is like that at the moment. My oft-repeated aim is to bring some method to this process - to allow starting with component values that are close to the optimum for a given system.
T&S and others have produced mathematical models which accurately predict the behaviour of electrodynamic drivers. It's a short step from there to incorporate the T-bass circuit in the model and predict the results. You have documented in another thread your use of such modelling techniques to predict T-bass operation. You've also stated that you don't believe such modelling techniques explain the way the T-bass works. This implies that the model did not show the results you had predicted. If the model does not match the theory, then at least one of them is wrong. My intention is to come up with a model that produces results that match the measurement of real-world speakers with T-bass.
As for ruining the thread, all of the information it contains is still there for anyone to read. Regarding my own participation, I have been back and reviewed the thread, starting with my first post here (5th March 2010). I believe I have been consistent, polite and open-minded, though by no means obsequious. I do note that much of what I have said in this post is also in my earlier posts. I will try harder to confine myself only to discussion of new points raised.
Regards,
Don.
Last edited:
You were not, as far as I can see.
BTW Its janneman, without the 'r'.
BTW-2 I referred John Curl to your ACE preamp in the Blowtorch II thread. John was looking for a way to implement variable cartridge loading, which you already used in the ACE 30 or 40 years ago.
John hasn't reacted yet though.
jd
It's like making rabbit stew - first, I have to catch a rabbit.
(Build an OB speaker.) Suitable 12 and 15 inch drivers are scarce and expensive in this country, and I have no space to use large OB speakers. But I have a plan, which I will document in a new thread.
Meanwhile, I'm making slow progress on a converter program that takes driver T-S parameters as input and produces a SPICE model of the driver with all the correct value components. I'm also investigating SPICE modellers - I'd like to find one that will accept a WAV file as the input function and produce a WAV file at the output. I can feed the model music and listen to the results on headphones.
(Graham also made a comment a while back that the problem fixed by T-bass doesn't occur with headphones. I thought so too, but now I wonder. One more thing to test...)
Actually, I believe Graham had a somewhat different purpose in mind when he started it. I'm quite happy to take the subthread covering the modeling of a T-bass to a different thread, so Graham can continue providing assistance to people here. When I have some significant progress to report, I'll start a new thread and point to it here.
Point taken. My purpose was to point out that, in my opinion, the perceived T-bass correction need is attributed to a signal that is unrealistic and doesn't occur in music. T-bass will surely modify the woofer response, which may even sound pleasant, but it makes it less accurate.
But I'll shut up now, I'm looking forward to your results anyway.
jd
Graham,
I've been using the T-Bass circuit for a few months now and I've found that it has taken while (50 hours or so) for it to break in and sound better. It really controls the Alpha 15 low end.
My question is this - I've noticed that I can hear midrange coming through the Alpha 15. (I will have to go back to see exactly what configuration I have used on the T-Bass.)
Should the T-Bass cut the Alpha 15 off at a certain frequency?
John
OB with B200/Alpha 15
I've been using the T-Bass circuit for a few months now and I've found that it has taken while (50 hours or so) for it to break in and sound better. It really controls the Alpha 15 low end.
My question is this - I've noticed that I can hear midrange coming through the Alpha 15. (I will have to go back to see exactly what configuration I have used on the T-Bass.)
Should the T-Bass cut the Alpha 15 off at a certain frequency?
John
OB with B200/Alpha 15
I am also trying to make improvised version of T-bass before I spring for trafos...but am somewhat confused with different cap nomenclature - in my country we call + - caps "bipolar" and type without + and - "unipolar".
So, if I understood properly two 470 uF caps are bipolar (+ -), yes?
And they should be connected in paralel (+ - ) and (- +), or maybe in series + (- - ) +...or - (+ +) - ? I keep digging out contradictory info...
Also, my mockup trafo (400 VA, 2x36) is not toroidal, is it just that toroidal trafos are better but not essential, or they have to be toroidal?
And, last question - I have three lines coming out of trafos (apart from primary lines which are not to be used) - red, black, red - what would be proper way of connecting them?
Thank you.
Oh, btw, Simon, your OB threads are one of the most usefull ever!
So, if I understood properly two 470 uF caps are bipolar (+ -), yes?
And they should be connected in paralel (+ - ) and (- +), or maybe in series + (- - ) +...or - (+ +) - ? I keep digging out contradictory info...
Also, my mockup trafo (400 VA, 2x36) is not toroidal, is it just that toroidal trafos are better but not essential, or they have to be toroidal?
And, last question - I have three lines coming out of trafos (apart from primary lines which are not to be used) - red, black, red - what would be proper way of connecting them?
Thank you.
Oh, btw, Simon, your OB threads are one of the most usefull ever!
Gamgee123,
I'm not sure which caps you're referring to but if you want to use polarised caps (normal electrolytics not designed for the audio signal-path) then you can connect them in series ie. +--+ (the minuses are joined - at least that's how I've done it once). The capacitance will halve so 2 x 470uF in series becomes 1 x 235uF. I think you can probably also parallel them in the way you said but I really am NOT sure! Best to stick with series connection, it does work for sure.
I think Graham said that toroids have the right resistance or inductance on their secondaries, so whether frame transformers work well may be down to those specifications, which could differ between transformers. I'm sorry I can't be of any more help in this department.
With your secondaries you have two ends and a middle ie. a centre-tap. Referring to Graham's diagram on page one (http://www.diyaudio.com/forums/full-range/130679-t-bass-drive-ob-lf-drivers.html) one red (either) is "0V", the other is "80V" and the black wire (centre-tap) is "40V". I do hope I have this correct!!!
With the diagram shown on page 1 it's a toroidal with a very typical configuration of 2 x secondary windings. If these are connected to each other it becomes, in practical terms, similar to your transformer with 3 distinct connection points.
ps - I'm glad the open baffle threads I started have proved useful, I'm very grateful to those people who posted and helped me so much as it's their knowledge and experience that makes the useful info - not the drivel I spew
I'm not sure which caps you're referring to but if you want to use polarised caps (normal electrolytics not designed for the audio signal-path) then you can connect them in series ie. +--+ (the minuses are joined - at least that's how I've done it once). The capacitance will halve so 2 x 470uF in series becomes 1 x 235uF. I think you can probably also parallel them in the way you said but I really am NOT sure! Best to stick with series connection, it does work for sure.
I think Graham said that toroids have the right resistance or inductance on their secondaries, so whether frame transformers work well may be down to those specifications, which could differ between transformers. I'm sorry I can't be of any more help in this department.
With your secondaries you have two ends and a middle ie. a centre-tap. Referring to Graham's diagram on page one (http://www.diyaudio.com/forums/full-range/130679-t-bass-drive-ob-lf-drivers.html) one red (either) is "0V", the other is "80V" and the black wire (centre-tap) is "40V". I do hope I have this correct!!!
With the diagram shown on page 1 it's a toroidal with a very typical configuration of 2 x secondary windings. If these are connected to each other it becomes, in practical terms, similar to your transformer with 3 distinct connection points.
ps - I'm glad the open baffle threads I started have proved useful, I'm very grateful to those people who posted and helped me so much as it's their knowledge and experience that makes the useful info - not the drivel I spew
First, thank you for answering so quickly.
I was referring to Grahams diagram on page one (it shows 2x 470 uF connected in parallel but reversed polarity). Wish me luck, I'm going to try this out...I'm sure I'll be back many more times to this thread...
Although I read and reread, my understanding of this circuit is vague at best...
BTW, what is your speaker measuring setup?
I was referring to Grahams diagram on page one (it shows 2x 470 uF connected in parallel but reversed polarity). Wish me luck, I'm going to try this out...I'm sure I'll be back many more times to this thread...
Although I read and reread, my understanding of this circuit is vague at best...
BTW, what is your speaker measuring setup?
I use a Panasonic electret capsule (WM-61A) soldered onto a "stick" of coaxial wire, which plugs into a preamp made from a kit (Mad About Sound mic pre-amp). This feeds my soundcard (SB X-Fi). The software I use is the free version of ARTA: ARTA Download
Simon
Simon
