I have modified the schematic a little bit.
http://img41.imageshack.us/img41/971/diagram1.png
http://img10.imageshack.us/img10/4710/diagram2.png
http://img41.imageshack.us/img41/971/diagram1.png
http://img10.imageshack.us/img10/4710/diagram2.png
Place the 0.1uF cap between pins 16 and 7, and the old 100pF cap place then between pins 15 and 16. This will be the best way, I think.
(please, do not swap uF and pF here )0.1uF and stabilized power supply is ok here.
In this project on Veroboard - NO.
You do not use any feedback, you do not need high precision measurements here, etc...Good Luck!
Place the 0.1uF cap between pins 16 and 7, and the old 100pF cap place then between pins 15 and 16. This will be the best way, I think.
Will Do. Thanks that simplifies things a lot more.
Alright that's good then. Also I am understanding you correctly right? Remove the copper track right after the inductors?
If they are connected _after_ the LC filter, then it is not a problem at all...
Yep they are
Yes!
You can even place the test point pins here, but directly near to the inductor leg (switching node). Also, the inductor's leg is a good test point pin too!
Yes!
You can even place the test point pins here, but directly near to the inductor leg (switching node). Also, the inductor's leg is a good test point pin too!
Alright I will leave 1 more hole for the test point from the leg of the inductor and wipe of the excess copper tracks remaining.
Actually where else should I remove copper tracks? If you see closely on the upper IR2110. I have another ground line and that
s where the Vss abd SD pin get pulled down to. Should I remove the copper track after these wires?
Same with the 12V as well. It just goes all the way.
Last edited:
Ground and power rail wires can stay as long, as it needed, I think... But the signal wires/traces should be as short as possible.
Also, I have not seen the copper side of your board yet... There is no post for it (maybe wrong picture?).
Last edited:
Ground and power rail wires can stay as long, as it needed, I think... But the signal wires/traces should be as short as possible.
Also, I have not seen the copper side of your board yet... There is no post for it (maybe wrong picture?).
Alright I will leave those tracks alone then.
I have not taken a picture of the copper side yet. I will post the picture tomorrow.
Ok my bad. Pafi was correct. My amplifier was drawing out 3A, whereas I had current Limited the power supply to just under 1A. So as you increased volume, it wanted more current. Hence why i saw the effect (I saw it again today). But today I received some help concerning it and it's all good now.
I have added the 100nF cap between pin 16 and ground (pin 7) and also the two 100nF caps between Vdd and Vss.
Again all are at the bottom of the board
I will post pics of the copper track later
I would like to ask from everyone's professional opinion. Is it wise to move to a Adapter that can convert AC voltage off the power sockets to 20V? Otherwise I will just continue using the DC power Supply I am using now.
I have added the 100nF cap between pin 16 and ground (pin 7) and also the two 100nF caps between Vdd and Vss.
Again all are at the bottom of the board
I will post pics of the copper track later
I would like to ask from everyone's professional opinion. Is it wise to move to a Adapter that can convert AC voltage off the power sockets to 20V? Otherwise I will just continue using the DC power Supply I am using now.
I have also removed the 2 small antenna's as you called it..lol
And also de-tracked the copper trace just after the inductor.
And here are the pictures of the bottom side.
Power side and Gate Driver
http://img14.imageshack.us/img14/2295/dsc00047er.jpg
XOR and TL494
http://img44.imageshack.us/img44/5278/dsc00048wt.jpg
And also de-tracked the copper trace just after the inductor.
And here are the pictures of the bottom side.
Power side and Gate Driver
http://img14.imageshack.us/img14/2295/dsc00047er.jpg
XOR and TL494
http://img44.imageshack.us/img44/5278/dsc00048wt.jpg
Last edited:
The class D output stage has a very bad power supply rejection ratio. This means, any fluctuations on the power supply (hum, etc) will go directly into the output. In half bridge configuration this will be proportional to the duty cycle. So, at zero output (50% duty cycle) the 50% of fluctuation amplitude will go to the output. In full bridge configuration these fluctuations will be rejected at 50% duty cycle, but will be increased then proportionally to the duty cycle change. So if there is no any feedback or correction applied (as in this project), then regulated PSU is very appreciated
The HF capacitors should have the possible shortest legs. Otherwise their effect is reduced dramatically...
The HF capacitors should have the possible shortest legs. Otherwise their effect is reduced dramatically...
Is the blue one on the TL494 the problem or the orange one (100pF) as well?
Alright I'll see how short i can cut the legs off then.
Last edited:
The class D output stage has a very bad power supply rejection ratio. This means, any fluctuations on the power supply (hum, etc) will go directly into the output. In half bridge configuration this will be proportional to the duty cycle. So, at zero output (50% duty cycle) the 50% of fluctuation amplitude will go to the output. In full bridge configuration these fluctuations will be rejected at 50% duty cycle, but will be increased then proportionally to the duty cycle change. So if there is no any feedback or correction applied (as in this project), then regulated PSU is very appreciated
Ahh bummer ok. I wanted to see if i can make it more user friendly but alright. I'll just stick with the DC power supply I am using now.
Also you know how i can test my amplifier in software?
Any recommendations?
I'm almost there now, just take a few THD graphs and FFT graphs and Hallelujah I am done (thank god).
You know how the professionals show graphs of distortion and decibel range and such. I would like to plot something similar just to at least show the teacher I tried.
Oh, you can do a lot to polish this
New PCB, add feedback, use toroids in output filter, and a lot more... But this is an simple education project, and if you are satisfied with the results, then it is done! About FFT: simply use the soundcard of your PC. I do not have any recommendations about the software, but you can look it in this thread, I think: http://www.diyaudio.com/forums/showthread.php?t=116590
Also, I have not ever done such measurements myself.
I have not sell any class D amp ever. All my expirience is a small testing PCB and a lot of not finished schematics for simulation in LTSpice Good luck!
Oh, you can do a lot to polish this
About FFT: simply use the soundcard of your PC. A do not have any recommendations about the software, but you can look it in this thread, I think: http://www.diyaudio.com/forums/showthread.php?t=116590
Also, I have not ever done such measurements myself.
Good luck!
Ok Dumb question (more dumb then usual). How exactly does this work?
Like how?
At the moment my input music is coming from the headphone outlet of the PC. If your saying use the sound card, I'm guessing more wires and such. Which means plugs I gotta get. Can you explain which plugs.
I'm guessing there will be like a input going to the amp which is the wire I am using (I think), but also a output which goes to something in the PC and this is where I am lost.
Oh, you can do a lot to polish this
I would like to attempt all of that, starting from scratch the proper way everyone has mentioned here.
I mean right back as PSPICE modeling. But in Holidays.
Leave the music input as it is now, and connect the output of the amp to the PC aux input as shown in the picture below.
Need to say, this can give you wrong result, because of residual. This will interfere with the soundcard ADC sampling frequency and distort the measurements. I have seen somewhere the AES recommendations about the FFT/THD analyzer input filter for class D amp parameter measurements. Ideally, this filter should be used here.
Also, the schematic below will measure the parameters of the single half bridge in your amp at the time. But this should be ok anyway.
Last Edit:
do not forget to connect some load to your amp (usually it is an 4-8 Ohm resistor).Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.