Rule 1: Dead time is the main cause of distortion in class D. Gate drive should have little (like 10ns) or no dead time (this is hard to achieve).
Rule 2: Keep switching supply currents confined in loops as small as possible and use SMD ceramic chip capacitors to get low supply impedance between 10Mhz and 100Mhz. Use PI filters to prevent the switching hash in the rails from propagating through PSU wiring.
Rule 3: Use RC snubbers to damp layout and transistor package parasitistics as required. Component Values have to be carefully chosen for maximum damping with minimum losses.
Rule 4: MOSFET gates have to be charged and discharged fast (and by fast I mean 20 to 200ns). Use the gate charge figures from MOSFET datasheets and the output current figures from driver datasheets to estimate gate charging times and choose a proper combination of MOSFETs and gate driver.
You have a lot of homework to do...
Just kidding...
Rule 2: Keep switching supply currents confined in loops as small as possible and use SMD ceramic chip capacitors to get low supply impedance between 10Mhz and 100Mhz. Use PI filters to prevent the switching hash in the rails from propagating through PSU wiring.
Rule 3: Use RC snubbers to damp layout and transistor package parasitistics as required. Component Values have to be carefully chosen for maximum damping with minimum losses.
Rule 4: MOSFET gates have to be charged and discharged fast (and by fast I mean 20 to 200ns). Use the gate charge figures from MOSFET datasheets and the output current figures from driver datasheets to estimate gate charging times and choose a proper combination of MOSFETs and gate driver.
You have a lot of homework to do...
Just kidding...
alfsch said:
You have set a great trap for him..........
Zox, you need to learn alot in the field of power switching basics......then only you can have a good understanding of Class-D...........If you are a true RF-Engineer then why haven't you displayed your RF skill set in designing your PCB layout......
which looks like a Linear Class-AB layout except groundplane...
Go on with this layout (or a better one) and IR2011 or IR2010. You may even end up with a good amplifier if you are the kind of person that learns from mistakes.
You will also need an oscilloscope capable of displaying some stuff up to 100Mhz. Do your best to borrow one if you don't already own it. This is a must for class D.
You will also need an oscilloscope capable of displaying some stuff up to 100Mhz. Do your best to borrow one if you don't already own it. This is a must for class D.
Eva,
What kind of st*pid suggestion you are doing...Of course that I am going to use oscilloscope..and guess what I do have one that works a lot higher than just 100MHz.Do I have to keep telling you that I work very intensive on RF Power stuff and that we have more than $500.000 (1/2 of million dollars) worth of equipment...god.Should I tell you that AR&D (Advance Research and Development) group that I work for has 80% people that used to work for ex-Motorola RF group.
Stop with your "funny" comments....
Regards
What kind of st*pid suggestion you are doing...Of course that I am going to use oscilloscope..and guess what I do have one that works a lot higher than just 100MHz.Do I have to keep telling you that I work very intensive on RF Power stuff and that we have more than $500.000 (1/2 of million dollars) worth of equipment...god.Should I tell you that AR&D (Advance Research and Development) group that I work for has 80% people that used to work for ex-Motorola RF group.
Stop with your "funny" comments....
Regards
I have been advised by other forum members to stop taking you serious and this is what I'm going to do. You are going to get little help with this attitude, son.
BTW: Why on earth are you sending me private messages telling me that your amplifier does not go well and that you need the schematics of my amplifier, if you already know everything about electronics? I already published my layouts, which is something that nobody does because they give far more information than a schematic, but only to people that have a minimum electronics background in order to know what to look for...
BTW: Why on earth are you sending me private messages telling me that your amplifier does not go well and that you need the schematics of my amplifier, if you already know everything about electronics? I already published my layouts, which is something that nobody does because they give far more information than a schematic, but only to people that have a minimum electronics background in order to know what to look for...
Hi zox. Here are some of my thoughts.
There are lots of nice people on this forum that will take time out of their day to offer their suggestions, but do not expect that you will receive help. Normally you would have to pay people for that...
Just because you say you're an RF engineer doesn't mean that we know you have or have access to proper test equipment. Eva wasn't being demeaning or rude, she was simply pointing out the fact that it is almost necessary to use high-bandwidth equipment so you can see what's really going on with your design.
As far as your amp design goes, since this is the first quick and dirty prototype you're at the point where you need to get it up and running and see how well it is operating. You don't need us for that!
zox2003 said:
There are lots of nice people on this forum that will take time out of their day to offer their suggestions, but do not expect that you will receive help. Normally you would have to pay people for that...
zox2003 said:
Just because you say you're an RF engineer doesn't mean that we know you have or have access to proper test equipment. Eva wasn't being demeaning or rude, she was simply pointing out the fact that it is almost necessary to use high-bandwidth equipment so you can see what's really going on with your design.
As far as your amp design goes, since this is the first quick and dirty prototype you're at the point where you need to get it up and running and see how well it is operating. You don't need us for that!
Well, what to say...
I received sample parts from IR (2011,2110 AND 2104).
Soldered 2104 on PCB,did some initial voltage measurements with dummy load (4 ohms) and everything looked fine.
Hooked speaker up on output...apply input signal and guess what...AMPLIFIER WORKS....Without input signal amplifier is dead silent.Currently amplifier is playing with no issues.
How does it sound???
I am sure that some people will be dissapointed with another fact (beside one that I am RF Engineer)--I am musician too (keyboard,bass player).My ear tells me that there is some distorsion in mid-highs and highs.Bass sound nice and tight (fast).Obviosly I have to admit that this IR2104 is not the best choice for MOSFET driver but still works and I think that for Subamplifier would be acceptable with some modification in driving circuit (adding compl. drivers).My next step will be building amplifier based on same concept using 2011 driver (half bridge) or LT1162 (received samples yesterday)-full bridge.
Also,I will do some measurements tommorow on this current module.
I will share all my class D projects here with everybody on forum even if some of the BIG people over here do not like that idea.I do not know how long will it take me to finish whole class D project but I promised that on end I will have complete amplifier with protection circuits and I will publish all documentation that you need to build one.NO SECRETS..
All parts to build one are available...
I will keep updating my post...
Regards,
Zox
I received sample parts from IR (2011,2110 AND 2104).
Soldered 2104 on PCB,did some initial voltage measurements with dummy load (4 ohms) and everything looked fine.
Hooked speaker up on output...apply input signal and guess what...AMPLIFIER WORKS....Without input signal amplifier is dead silent.Currently amplifier is playing with no issues.
How does it sound???
I am sure that some people will be dissapointed with another fact (beside one that I am RF Engineer)--I am musician too (keyboard,bass player).My ear tells me that there is some distorsion in mid-highs and highs.Bass sound nice and tight (fast).Obviosly I have to admit that this IR2104 is not the best choice for MOSFET driver but still works and I think that for Subamplifier would be acceptable with some modification in driving circuit (adding compl. drivers).My next step will be building amplifier based on same concept using 2011 driver (half bridge) or LT1162 (received samples yesterday)-full bridge.
Also,I will do some measurements tommorow on this current module.
I will share all my class D projects here with everybody on forum even if some of the BIG people over here do not like that idea.I do not know how long will it take me to finish whole class D project but I promised that on end I will have complete amplifier with protection circuits and I will publish all documentation that you need to build one.NO SECRETS..
All parts to build one are available...
I will keep updating my post...
Regards,
Zox
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.