Choco, you talk about using a full bridge, from +/- 55V.
In a linear amp, a full bridge means that four transistors take the heat instead of two.
But in class D, I don't think a full bridge gives you any advantage, as far as ability to dissipate heat.
Of course, it gives you a bigger voltage swing at the output.
But couldn't you just use two transitors and higher voltages, especially at 2 Ohms.
Am I missing something?
In a linear amp, a full bridge means that four transistors take the heat instead of two.
But in class D, I don't think a full bridge gives you any advantage, as far as ability to dissipate heat.
Of course, it gives you a bigger voltage swing at the output.
But couldn't you just use two transitors and higher voltages, especially at 2 Ohms.
Am I missing something?
Choco, I understand all you are saying about BJT.
Most of those issues are relaxed if the frequency is lower.
This is just the first concept phase.
See what would be acceptable levels of switching signal bleed through at 20khz to 30khz.
Some folks have even looked at IGBT, and some other new power devices. Some are not even self commutating. So there would be more complexity. But those doors don't open unless the frequency is low.
Right now I just want to look at physiological aspects of 20khz to 30khz continuously.
Most of those issues are relaxed if the frequency is lower.
This is just the first concept phase.
See what would be acceptable levels of switching signal bleed through at 20khz to 30khz.
Some folks have even looked at IGBT, and some other new power devices. Some are not even self commutating. So there would be more complexity. But those doors don't open unless the frequency is low.
Right now I just want to look at physiological aspects of 20khz to 30khz continuously.
zenmasterbrian said:
Basically you could of course use a half bridge running from +/-110V.
But MosFet properties are getting poorer for higher voltage types.
Rds ON is increasing roughly by square of the voltage rating if you keep the chip area the same... A full bridge of course is still suffering from the same: You use already double chip area but still have two times the Rds ON in the path. But there are more points:
- Driving heavy MosFets with some ns is tough.
- Properties of the body diode are becoming really unfunny!!!
- PSRR: In generall SM amps are critical in power supply rejection, because the rail voltage is directly feeding into the output.
In a halfbridge only the symmetrical ripple of the rails is more or less eliminated. Means if the pos rail is higher than the negative rail, then open loop gain for positive signals is higher than for negative signals.
The full bridge is eliminating even unsymetrical rail ripple, because open loop gain is influenced by the difference from pos rail to neg rail for both half waves.
In the moment I am playing around with the IRS 20954 (half bridge driver) and IRFB52N15D (MOSFET: 150V/60A/32mOhms) and after some R&D work now I am able to adjust the dead time somewhere
short before crossconduction, means it can be anything between zero and 10ns... Well, I think that's no so bad, even if our gurus like Lars tell me that 10ns are King Kong's feet. I guess I cannot compete to his ns precision.
Of course all timing and distorsion issues are relaxing with lower frequencies. But already at 20kHz, it will become hard to get really good results with heavy bipolars.... I mean you would need at least something like 250V/40A types for your 1kW halfbridge.
My proposal:
Start to play around without much costs, just some BD249C (100V/15A) and +/-15V ... +/-40V rails and currents below 15A and see if you can get the small BJTs to sufficient switching performance. If yes, you can step up to higher power by using the results and may be found tweaks.
Safety
... in my last post I did not say anything about safety.
But if you really consider 1kW from a halfbridge, you must consider that you are handling voltages, which can easily kill humans.
You will need at least rails around +/-70V, if we consider voltage sagging, you would probably need something like +/-80V..90V (unloaded) to achieve honest 1kW into 2 Ohms. Take care, you can touch inbetween 180V.
Also take care that your amp may not kill your kids, when they are searching the rabbit, which might sit in the most hidden corner- trying to escape the 25kHz tweeting....
... in my last post I did not say anything about safety.
But if you really consider 1kW from a halfbridge, you must consider that you are handling voltages, which can easily kill humans.
You will need at least rails around +/-70V, if we consider voltage sagging, you would probably need something like +/-80V..90V (unloaded) to achieve honest 1kW into 2 Ohms. Take care, you can touch inbetween 180V.
Also take care that your amp may not kill your kids, when they are searching the rabbit, which might sit in the most hidden corner- trying to escape the 25kHz tweeting....
Choco, about the voltages, this is going to be a powered speaker. So there won't be contact with the speaker wires.
Have you identified good bipolars to consider for something like this? Say with 23.5khz switching frequency.
There are also people who have talked about IGBT, GTO, and even SCRs. Of course some of these require induced communtation.
Choco, I can see that you have really thought through some of the issues here. That is what I am starting to do.
Have you identified good bipolars to consider for something like this? Say with 23.5khz switching frequency.
There are also people who have talked about IGBT, GTO, and even SCRs. Of course some of these require induced communtation.
Choco, I can see that you have really thought through some of the issues here. That is what I am starting to do.
zenmasterbrian said:
you would be hard pressed to find a modern BJT that cannot function adequately at that kind of frequency.
I have not searched for suitable types.
But I am not so optimistic as fokker.
I would agree in most SMPS applications, where halfbridges are allowed to have long dead times. But for class D with low distorsions it is more difficult. I am mostly concerned about the storage time.
The storage time of heavy BJTs is long, means several us. You can reduce it by giving high negative base drive currents for turn OFF. Unfortunately the storage time is also depending on the overdrive factor. If you have a constant base drive then the resulting storage time will also depend on the load current. And temperature will increase the storage time by 30%...50% if you compare the behaviour at 100C vs. 25C.
Simply pick any power BJT and use it to switch ON and OFF a resistor load. Drive it with a signal generator and observe the delay between Ib and Ic. Or if you are in trouble with proper current measurement you can measure the time difference between the falling edge of the drive signal and the rising edge of Uce. Of course also there is a similar turn ON delay, usually this is much shorter, than turn OFF delay. Play around with this and gain some feeling about the values. Then go ahead with some drive buffer to allow high base drives or have a look to some of already existing standard base drive circuits. Play with it and see if you can keep the variance of delay times low no matter if you heat up and/or change the load.
I would guess that you can be lucky if you are able to drive a 250V/40A BJT in a way that the resulting turn OFF delay is about 1.5us at 25C and 2us at 100C. Which is already a variance of 500ns and does not take load into account. From my perspective short dead times for class D half bridges are really difficult to realize in a proper manner with BJTs.
On the other hand.... We have some people here, who know better about BJTs, than I do. So I am not the right one to ask.
But I am not so optimistic as fokker.
I would agree in most SMPS applications, where halfbridges are allowed to have long dead times. But for class D with low distorsions it is more difficult. I am mostly concerned about the storage time.
The storage time of heavy BJTs is long, means several us. You can reduce it by giving high negative base drive currents for turn OFF. Unfortunately the storage time is also depending on the overdrive factor. If you have a constant base drive then the resulting storage time will also depend on the load current. And temperature will increase the storage time by 30%...50% if you compare the behaviour at 100C vs. 25C.
Simply pick any power BJT and use it to switch ON and OFF a resistor load. Drive it with a signal generator and observe the delay between Ib and Ic. Or if you are in trouble with proper current measurement you can measure the time difference between the falling edge of the drive signal and the rising edge of Uce. Of course also there is a similar turn ON delay, usually this is much shorter, than turn OFF delay. Play around with this and gain some feeling about the values. Then go ahead with some drive buffer to allow high base drives or have a look to some of already existing standard base drive circuits. Play with it and see if you can keep the variance of delay times low no matter if you heat up and/or change the load.
I would guess that you can be lucky if you are able to drive a 250V/40A BJT in a way that the resulting turn OFF delay is about 1.5us at 25C and 2us at 100C. Which is already a variance of 500ns and does not take load into account. From my perspective short dead times for class D half bridges are really difficult to realize in a proper manner with BJTs.
On the other hand.... We have some people here, who know better about BJTs, than I do. So I am not the right one to ask.
Choco,
There are clearly many dimensions to this.
Eventually I will want to start compiling a data collection for the most capable output devices, BJT, MOS, as well as the other technologies.
Lets let this thread sit now, with the open question about the physiological and environmental limits of 23.5khz exposure.
During the 1970's there were class D amps that ran at these frequencies. Likewise there were, and sometimes still are, SMPS and inverters that run this low. There are also other types of ultrasonic appliances that run in the 20khz to 30khz range.
So there must have developed some accepted conventions about what level of low ultrasonic exposure is acceptable, when it is continuous.
It must not only be not harmful, it should be completely imperceptible.
Lets let this thread sit, and see if anyone has any links to address this.
There are clearly many dimensions to this.
Eventually I will want to start compiling a data collection for the most capable output devices, BJT, MOS, as well as the other technologies.
Lets let this thread sit now, with the open question about the physiological and environmental limits of 23.5khz exposure.
During the 1970's there were class D amps that ran at these frequencies. Likewise there were, and sometimes still are, SMPS and inverters that run this low. There are also other types of ultrasonic appliances that run in the 20khz to 30khz range.
So there must have developed some accepted conventions about what level of low ultrasonic exposure is acceptable, when it is continuous.
It must not only be not harmful, it should be completely imperceptible.
Lets let this thread sit, and see if anyone has any links to address this.
A new thread to talk about the dead time or crossover distortion issue.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=84403
For this thread, just let it sit with the open question about the physiological and environmental limits of 23.5khz exposure.
Eventually one of us will come up with some more links that speak to this.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=84403
For this thread, just let it sit with the open question about the physiological and environmental limits of 23.5khz exposure.
Eventually one of us will come up with some more links that speak to this.
Hi zenmasterbrian,
Enough. You can not apply a filter on an open public thread. This is an open discussion and all members are free to contribute their thoughts on the matter. You can't pick and choose what they post, or who posts. You can put a member or members on your ignore list. That is the reality of public forums.
We do not allow members to attack other members personally, or bait them constantly. I must point out that the posts were helpful in the beginning, and from some extremely experienced people on those topics. Please try and accept all posts, both good and bad. There is no way we are going to muzzle the bulk of the membership.
Also accept that there are certain topics that will invite a spirited discussion. This is one of them. I believe the proper course of action at this point would be for you to build your prototype and solve the issues as they come up. You will find some very helpful members to guide you through specific issues.
At the end of the exercise, you will know more from the experience and can report with some authority. I don't think arguing about a thought experiment will get you very much further. No prototype works perfectly the first time out, so build it. Make mistakes and fix them. That's what we all do.
-Chris
Okay.
Enough. You can not apply a filter on an open public thread. This is an open discussion and all members are free to contribute their thoughts on the matter. You can't pick and choose what they post, or who posts. You can put a member or members on your ignore list. That is the reality of public forums.
We do not allow members to attack other members personally, or bait them constantly. I must point out that the posts were helpful in the beginning, and from some extremely experienced people on those topics. Please try and accept all posts, both good and bad. There is no way we are going to muzzle the bulk of the membership.
Also accept that there are certain topics that will invite a spirited discussion. This is one of them. I believe the proper course of action at this point would be for you to build your prototype and solve the issues as they come up. You will find some very helpful members to guide you through specific issues.
At the end of the exercise, you will know more from the experience and can report with some authority. I don't think arguing about a thought experiment will get you very much further. No prototype works perfectly the first time out, so build it. Make mistakes and fix them. That's what we all do.
-Chris
This is indeed an open public forum, and it is also broken into topics. There are 1200 other topics.
I agree that a great deal of useful information was communicated.
Where people wanted to challenge the premises of what I am trying to do, I did discuss it with them. But I also invited them to narrow it back to the topic.
No one is forcing people to post here.
I have tollerated continual baiting, and even badgering. When people have done this, I have held to a more narrow topic def'n.
Such baiting and badgering is entirely inappropriate. It should be deleted, because it mucks up the thread for future readers. It discourages other people from making topical posts. It makes it hard for readers to understand what is going on.
This topic seems to have provoked controversy, but there is no reason for that. Its only because some people believe that there is only one right way, and they know it.
When I was a teenager, I used to build things without much background information.
I am not a teenager anymore. I have a higher expectation for the quality of the result.
If its not possible to have respectful topical discussion, then this board does serve little purpose.
**********
Lets let this thread sit now, with the open question about the physiological and environmental limits of 23.5khz exposure.
Eventually one of us will come up with some relevant material.
I agree that a great deal of useful information was communicated.
Where people wanted to challenge the premises of what I am trying to do, I did discuss it with them. But I also invited them to narrow it back to the topic.
No one is forcing people to post here.
I have tollerated continual baiting, and even badgering. When people have done this, I have held to a more narrow topic def'n.
Such baiting and badgering is entirely inappropriate. It should be deleted, because it mucks up the thread for future readers. It discourages other people from making topical posts. It makes it hard for readers to understand what is going on.
This topic seems to have provoked controversy, but there is no reason for that. Its only because some people believe that there is only one right way, and they know it.
When I was a teenager, I used to build things without much background information.
I am not a teenager anymore. I have a higher expectation for the quality of the result.
If its not possible to have respectful topical discussion, then this board does serve little purpose.
**********
Lets let this thread sit now, with the open question about the physiological and environmental limits of 23.5khz exposure.
Eventually one of us will come up with some relevant material.
Zen,
The mods here are pretty cool dudes... DON'T wang on 'em. Some of them have taken their personal time to help me with non-forum issues. I have helped some of them a time or two. Do you REALIZE what they get paid? They do this because they like helping people, learning, and teaching. It's also a social group where people don't have to drive drunk just to shoot the **** with the boys (and Evas), made better because the "boys" are peers. It is also OK to: poke in the ribs, punch in the shoulder, and pour beer on the next guy.
The only people who will be reading this thread in the future will be looking for comic value. As far as mucking things up is concerned, you're getting plenty of bandwidth here... mostly the same message... and ALOT of BIGTIME crybaby stuff.
No, it because of all the ways that have been mentioned, you insist on the wrong one. Case in point, I have posted at least 3 relevant links to the physiological knowledge you seek. Perhaps you are looking for more; but your logic is failing you. If you have read and believe what has been provided you thus far, you would extrapolate a bit, and realize that the effects of ultrasonic at an SPL pf 20 dB aren't worth worrying about. Class D's assessment of your willingness to do your own homework is apparently correct.
A lot of the foreign posters work overtime... place an extra effort to speak in the best English they can... time and effort. I, and others, have spent more than just a minute or two finding info for you to ignore, only because you asked.
In terms of disrespect to the forum... you have the monopoly at this point.
🙁
The mods here are pretty cool dudes... DON'T wang on 'em. Some of them have taken their personal time to help me with non-forum issues. I have helped some of them a time or two. Do you REALIZE what they get paid? They do this because they like helping people, learning, and teaching. It's also a social group where people don't have to drive drunk just to shoot the **** with the boys (and Evas), made better because the "boys" are peers. It is also OK to: poke in the ribs, punch in the shoulder, and pour beer on the next guy.
The only people who will be reading this thread in the future will be looking for comic value. As far as mucking things up is concerned, you're getting plenty of bandwidth here... mostly the same message... and ALOT of BIGTIME crybaby stuff.
No, it because of all the ways that have been mentioned, you insist on the wrong one. Case in point, I have posted at least 3 relevant links to the physiological knowledge you seek. Perhaps you are looking for more; but your logic is failing you. If you have read and believe what has been provided you thus far, you would extrapolate a bit, and realize that the effects of ultrasonic at an SPL pf 20 dB aren't worth worrying about. Class D's assessment of your willingness to do your own homework is apparently correct.
A lot of the foreign posters work overtime... place an extra effort to speak in the best English they can... time and effort. I, and others, have spent more than just a minute or two finding info for you to ignore, only because you asked.
In terms of disrespect to the forum... you have the monopoly at this point.
🙁
Thanks to poobah and others who have come up with relevant links.
Lets let this thread sit now, with the open question about the physiological and environmental limits of 23.5khz exposure.
Eventually one of us might come up with some additional relevant material.
😎
Lets let this thread sit now, with the open question about the physiological and environmental limits of 23.5khz exposure.
Eventually one of us might come up with some additional relevant material.
😎
poobah said:
Post of the year sticky award. We ought to have a "classics" thread don't you think?
Let's start a vote for that feature, sometimes texas is far too punitive, topically.
To get such a reply as below, I begin to think the forum is hacked.
ZenmasterBrian,
A forum is a community,of real people.
When someone here is googling for you half an hour because you're too lazy to do it yourself,
the least you can do is show him you've done something with what he GAVE you.
If someone goes half across town to a library trying to find a book for someone he/she barely knows,
he/she would like to hear a little more than "thanks for the info, i'll read it".
This thread is not your private mansion, where people come just to deliver their books to you and leave empty-handed.
You come here looking for knowledge, yet you seem mostly content with the "solution" you came up with BEFORE you had the knowledge.
I know i can be DARN stubborn, but you take the cake !
If you don't show some progress, or at least that you're TRYING to understand , what is the use in providing you with more knowledge ?
Also, trying to help other people goes a long way here.
It's a two-way street here duud.
btw: my SACD-player has a frequency-range that far exceeds 23.5 khz. Should i be worried about my pets ? 🙄
Perhaps it's better if i leave the grille-cloths on the speakers
Now let's just let this thread sit and wait for another highly redundant remark from ZenmasterB
Klaas
A forum is a community,of real people.
When someone here is googling for you half an hour because you're too lazy to do it yourself,
the least you can do is show him you've done something with what he GAVE you.
If someone goes half across town to a library trying to find a book for someone he/she barely knows,
he/she would like to hear a little more than "thanks for the info, i'll read it".
This thread is not your private mansion, where people come just to deliver their books to you and leave empty-handed.
You come here looking for knowledge, yet you seem mostly content with the "solution" you came up with BEFORE you had the knowledge.
I know i can be DARN stubborn, but you take the cake !
If you don't show some progress, or at least that you're TRYING to understand , what is the use in providing you with more knowledge ?
Also, trying to help other people goes a long way here.
It's a two-way street here duud.
btw: my SACD-player has a frequency-range that far exceeds 23.5 khz. Should i be worried about my pets ? 🙄
Perhaps it's better if i leave the grille-cloths on the speakers
Now let's just let this thread sit and wait for another highly redundant remark from ZenmasterB
Klaas
confucius say, one child, whhhhhhoooooooo communitaww.
Don't his posts remind you of the movie 1984?
Don't his posts remind you of the movie 1984?
to save zenmasterbrian from having to type it again... here it is. 😀 
I have a feeling it would be posted again so.......
I have a feeling it would be posted again so.......
zenmasterbrian said:
nah. I'm too busy to look up for the meta post comment thing just to post it again. but I guess it would go something along the lines......
"classd4sure, you are not required to post meta on this thread".... and then.......
but I guess it would go something along the lines......"classd4sure, you are not required to post meta on this thread".... and then.......
zenmasterbrian said:
- Status
- Not open for further replies.