Ok, I'll build it.
But...I'll need some tips and tricks on how to build and set an amplifier like this one.
I'm not a newbie, but I'm an analogic builder, and my experience on class D is only Tripath 'plug'n play' chips.
So, can you please tell me a little more about the working operation?
In particular, how can I modify Rt/Ct ratio?
I don't see any Ct nor Rt on the main schematic...
Which power fet should I choose?
Is +90V DC smps ok, or maybe better??
Txia for spending your time here.
p.s: is this a classic Class D amp? (so I can study in deep Class D operation...)
Yes,
I agree with you.
So this means I can modify switching frequency later too.
Very good.
I said +90V, but I don't know which is the best voltage to choose for the best audio quality.
Anyway, can I go with something like this?
It's only an example.
Can I choose 50V/60V/70V ?
Only single ended psu is needed?
I agree with you.
So this means I can modify switching frequency later too.
Very good.
I said +90V, but I don't know which is the best voltage to choose for the best audio quality.
Anyway, can I go with something like this?
It's only an example.
Can I choose 50V/60V/70V ?
Only single ended psu is needed?
Edit:
I don't understand:
the main schematic on the first post seems to have a single power supply (+V/ground);
the PCB has dual power supply (+V/-V/ground).
And also the fact that 12V external supply is referred to -V (and not to ground) let me think there is a dual power supply, but the schematic shows only a single power supply.
Where's the truth?
I don't understand:
the main schematic on the first post seems to have a single power supply (+V/ground);
the PCB has dual power supply (+V/-V/ground).
And also the fact that 12V external supply is referred to -V (and not to ground) let me think there is a dual power supply, but the schematic shows only a single power supply.
Where's the truth?
Ok,
I saw datasheets and learned something more about class D.
So, before I start to collect the stuff, I would like to know if this project is ok for hifi, or only PA.
My goal is a very hi fidelity and hi quality ampli, 100Wrms is good enough.
Don't know if this is the right way...
I would like to start a diy class D hi quality full range ampli, but first is better to start with a collaudate design.
I saw datasheets and learned something more about class D.
So, before I start to collect the stuff, I would like to know if this project is ok for hifi, or only PA.
My goal is a very hi fidelity and hi quality ampli, 100Wrms is good enough.
Don't know if this is the right way...
I would like to start a diy class D hi quality full range ampli, but first is better to start with a collaudate design.
This is pure class D.It can be described as synchronous buck converter.
It uses single psu,and any voltage from 0V to 200V can be used.Only power and output volume depends of power supply.You can chenge volume with changing power supply voltage without any modification,and changing volume at input.This is because fer example when duty cycle is say 60%,on 5V it will be 3V at out,but on 50V it will be 30V.12V is referenced to negative supply.You can use center taped,but center tap of transformer can't be connected with speaker because SG IC maybe have temp drift or another thing,but you can try.Set voltage at 0V and play music about 10 min's.Monitor voltage at output in any 2 minutes without signal.If voltage after 10min's remain max. +/-100mV you can use dual psu.
Download SG IC datasheet and see what's CT and RT pin's.
If you have another Q's post here,I will respond as fast as I can.
Cheers
It uses single psu,and any voltage from 0V to 200V can be used.Only power and output volume depends of power supply.You can chenge volume with changing power supply voltage without any modification,and changing volume at input.This is because fer example when duty cycle is say 60%,on 5V it will be 3V at out,but on 50V it will be 30V.12V is referenced to negative supply.You can use center taped,but center tap of transformer can't be connected with speaker because SG IC maybe have temp drift or another thing,but you can try.Set voltage at 0V and play music about 10 min's.Monitor voltage at output in any 2 minutes without signal.If voltage after 10min's remain max. +/-100mV you can use dual psu.
Download SG IC datasheet and see what's CT and RT pin's.
If you have another Q's post here,I will respond as fast as I can.
Cheers
Yes,can be used for hi-fi,set frequency on 300kHz,I described this in few posts up.For 100W on 4 ohms use single 70V supply.
Ok, I studied...
so, I want R12=5K instead of 10K to obtain an oscillating frequency of 300KHz, and C1 remains the same (1nF); so it maintains the best (fast) discharge time (in fact Rd=0).
I found that I have to go with the same Ct and a smaller Rt. It's wrong?
so, I want R12=5K instead of 10K to obtain an oscillating frequency of 300KHz, and C1 remains the same (1nF); so it maintains the best (fast) discharge time (in fact Rd=0).
I found that I have to go with the same Ct and a smaller Rt. It's wrong?
Last edited:
No,it's better to use 10k and 470p.
Because big Ct will increase dead time,and output max. duty will be lower-smaller voltage at full input.
Thank you again.
I now know almost all I need to build and understand this project.
Generally speaking, we say this is classical class D.
And classical class D is affected (was?) by low damping factor, and in general the low point is the output filter, composed by Loutput, Coutput and Loudspeaker, that we know is not a constant impedance on the full frequency range and therefore the distorsion figure is not constant on the full frequency range.
I knew that to overcome this limit is necessary to apply some quantity of negative feedback, and some elaborate algorithms: I mean Hypex and so on, that are immune to the loudspeakers impedance variability.
This project is a simple amply (I like minimalism on audio): how it performs under this aspect?
I now know almost all I need to build and understand this project.
Generally speaking, we say this is classical class D.
And classical class D is affected (was?) by low damping factor, and in general the low point is the output filter, composed by Loutput, Coutput and Loudspeaker, that we know is not a constant impedance on the full frequency range and therefore the distorsion figure is not constant on the full frequency range.
I knew that to overcome this limit is necessary to apply some quantity of negative feedback, and some elaborate algorithms: I mean Hypex and so on, that are immune to the loudspeakers impedance variability.
This project is a simple amply (I like minimalism on audio): how it performs under this aspect?
Yes,low damping factor is because this is a synchronous buck/boost.When voltage at speaker is higher than average pwm voltage,it work as boost and back energy to power supply.Feedback can be easily added,but first try without it.It's immune too,I use it from 0,5 to 16 ohm speakers (series-parallel combination) and quality is same (as I heard).I like minimalism at all.In few weeks I will try mag amp amplifier.This is very simple,and don't use diodes and rectifiers on smps out.It drive out directly.
Ok, I'll try it same as it is now.
I'll only change the swhitching frquency to 300KHz.
What about power supply?
Can I employ a linear 80V/300VA single ended psu, or the buck/boost phenomena can damage it?
I didn't know mag amp; fast research on google tells me a keyword: Bob Carver (ok, two keys...).
Do you refer to a magnetic power supply or to an actual amplifier?
EDIT: from here:
"Magnetic Amplifiers:
Although called a magnetic amplifier, this application really
uses an inductive element as a controlled switch. A mag
amp is a coil of wire wound on a core with a relatively
square B-H characteristic."
Last edited:
Mag amp isn't bad as doctormord say.I tried it without feedback,and sound is ok for me.
You can use any PSU,boost can't increase voltage on electrolytes because speaker reactive energy is small to do this.As I say,first try on 150khz.I make mag amp in halfbridge and pushpull,and it don't require PSU cap's.SMPS directly power it.But this is for sub and mid's because sw. freq. is cca. 35khz
You can use any PSU,boost can't increase voltage on electrolytes because speaker reactive energy is small to do this.As I say,first try on 150khz.I make mag amp in halfbridge and pushpull,and it don't require PSU cap's.SMPS directly power it.But this is for sub and mid's because sw. freq. is cca. 35khz
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.