I also tried that 6 years ago before I publish my tl494 thread but upon checking in my scope, I noticed that the emitter follower is more longer in width than the common emitter. So, I did not try to listen to both of its output independently. Its almost good to
Implement it but they are not symmetrical as far as I can remember. So I opted for inveter as the signal splitter
Implement it but they are not symmetrical as far as I can remember. So I opted for inveter as the signal splitter
These tricks with one output transistor as common emitter and the other one as emitter-follower will not work. This IC is designed for push-pull or half-brideg SMPS operation. There is a toggle- FF inside this IC that serves to steer one transistoer during one phase and the other one during the other one. In order to be used for class-d audio they would have to be combined in wired-or fashion. There is absolutely no way (at least not known to me) around that. So your only choice is to have an additional IC that acts as an inverter between the TL 494 and an IR2110.
Regards
Charles
Regards
Charles
This is what i've implemented myself personally, before my teacher said go PWM chip.
I used Circuitmaker to simulate results and followed it up with a experimental result. The op-amp is crap, i get a sine wave instead of a triangle at 100KHz because of its slew rate, although its better at 50KHz.
Here's what I've done personally and do understand
Simulations:
http://img198.imageshack.us/i/crap2.png/
just better point out, I randomly used the MAX942 because i couldnt find a model of LM361N comparator in circuitmaker. But the comparator i use experimentally is the LM361N, which gives me complementary outputs.
Experimental:
http://img198.imageshack.us/i/tek0012.jpg/
I get ringing effect cause my wiring techniques are still clumsy. Had long wires connected everywhere, which led to parasitic inductance. I'll fix it up sooner or later.
I used Circuitmaker to simulate results and followed it up with a experimental result. The op-amp is crap, i get a sine wave instead of a triangle at 100KHz because of its slew rate, although its better at 50KHz.
Here's what I've done personally and do understand
Simulations:
http://img198.imageshack.us/i/crap2.png/
just better point out, I randomly used the MAX942 because i couldnt find a model of LM361N comparator in circuitmaker. But the comparator i use experimentally is the LM361N, which gives me complementary outputs.
Experimental:
http://img198.imageshack.us/i/tek0012.jpg/
I get ringing effect cause my wiring techniques are still clumsy. Had long wires connected everywhere, which led to parasitic inductance. I'll fix it up sooner or later.
Isn't single ended input and wired-OR fashion the same thing.
You connect them all to some point, such that they are then pulled up by a resistor. I'm doing that right now with C1 and C2
http://zone.ni.com/devzone/cda/tut/p/id/3544
You connect them all to some point, such that they are then pulled up by a resistor. I'm doing that right now with C1 and C2
http://zone.ni.com/devzone/cda/tut/p/id/3544
Pafi said:
Only except my teacher is being an ***, by telling someone who is new to this to go and use complex chips. I could strangle him.
It was so easy to verify my results using circuitmaker following my own method before implementing it on breadboard.
What do you mean by check pin 13?
It's either pulled down to ground for single ended output or pulled to Vref for push-pull.
Hi circuitcity,
Can you please show me in your scope about that 1 volt offset? Do you mean the complementary outputs are not symmetrical? If one output has higher amplitude but both are same in width then this is ok. The input buffer of your fet driver will take care of it. But if one has more wider than the other then this is not good.
You can use 1N914 in parallel with the gate resistor but if it gets warmer then use two of it in parallel depending in the gate capacitance of your fet. Or you can use fast scotkey diode like 1N5819. I just used 1N914 in my previous experiments and it works well but I have smaller power fet though.
Cheers
Can you please show me in your scope about that 1 volt offset? Do you mean the complementary outputs are not symmetrical? If one output has higher amplitude but both are same in width then this is ok. The input buffer of your fet driver will take care of it. But if one has more wider than the other then this is not good.
You can use 1N914 in parallel with the gate resistor but if it gets warmer then use two of it in parallel depending in the gate capacitance of your fet. Or you can use fast scotkey diode like 1N5819. I just used 1N914 in my previous experiments and it works well but I have smaller power fet though.
Cheers
Just one question to ask you Ledmania. When you applied the music to observe the sound across the speaker. Did you reference the music signal from the MP3 player to the global ground or to the mid-point?
Because the ground of the speaker is respect to the mid-point of the two supplies. I believe my mistake today was that i set the ground for the music signal as the global ground and not the mid-point.
Because the ground of the speaker is respect to the mid-point of the two supplies. I believe my mistake today was that i set the ground for the music signal as the global ground and not the mid-point.
Everything upto the IR2110 Gate driver has been referred to one ground.
But for the output MOSFETS and speakers, what I've gone and done is taken that Mid-point between the two supplies as another ground and am wondering how that is affecting the circuit now.
Or is my grounding technique completely wrong?
But for the output MOSFETS and speakers, what I've gone and done is taken that Mid-point between the two supplies as another ground and am wondering how that is affecting the circuit now.
Or is my grounding technique completely wrong?
So, you are already at the fet driver stage then? Please be careful for for cross conduction. Put at least 1 ohm 1watt resistor in series
with your main DC supply and sample the scope probe accross to that resistor without music and observe the pulses signal that appear accross that resistor. By using ohms law, you will get the current flow to it. If you get higher than 750ma then your deadtime
is not enough. should be at lease less than 200ma. At 750ma
or more, you will see sharp transient signal in the order of usec in the leading and trailing edges of the pulse. this is an indications of shoot through.
cheers
with your main DC supply and sample the scope probe accross to that resistor without music and observe the pulses signal that appear accross that resistor. By using ohms law, you will get the current flow to it. If you get higher than 750ma then your deadtime
is not enough. should be at lease less than 200ma. At 750ma
or more, you will see sharp transient signal in the order of usec in the leading and trailing edges of the pulse. this is an indications of shoot through.
cheers
ledmania said:
Allright thanks. I'll try with 1ohm, 1 watt resistor between the vsupply and Drain of High side mosfet and -Vsupply and source of low side mosfet.
Allthough I'm pretty sure my grounding technique is wrong. I'll have to sort out my grounding method first.
I have 2 breadboards.
On the one breadboard I have a red and blue rail for +12 volts and the negative terminal goes to ground.
This supply powers the TL494, Schmitt trigger and the gate driver.
Basically all my electronics circuits
On the second breadboard I have the POWER MOSFETS. With the +10V, -10V supplies and the mid-point ground of those.
I did reference the fixed sine wave to ground (global and not the mid-point) and the speaker to the Mid-point ground.
I can hear the single frequency i set the generator to but my graphs are completely messed up when I probe the mid-point of those two mosfets.
The suggestion given to me was, my grounding technique is wrong and more or less I've gone and shorted something along the way, which is why i hear the fixed freq but my graph is messed up.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.