วีดีโอ

github.com/SteveDSP/LTSpice

Hopefully that will help. I've included some others too.

@@stevenfenton3319 thanks a lot sir!

@mikegomez70 Glad it was of use.

No, it just works with transient analysis. If you're looking to do THD or spectral analysis, you can use the .four command.

What is the problem? Can you elaborate please?

@@stevenfenton3319 I'm trying to play wav file, like audio file i create circuit for it every time they show after run simulation " number of bits in a wave must from 1 to 32" now how i can solve it.... my file is audio i convert it in different kbs mbs but same problem

@@stevenfenton3319 and can we use universal OPamp in place of this opamp which you used in your circuit?

​@zeeshan.khan0 make sure your input wav file is valid, in terms of bit depth and sampling rate. 16but, 44.1k is CD quality. Specify this 16 and the 44 1 in the output script. You can see how I do this in the video.

​@zeeshan.khan0 if you're simulating my circuit, most general purpose or audio opamos will work.

Yes. Can be useful of course for switching, as you've highlighted.

Absolutely.

Glad it helped.

Glad it was helpful. I will add more when I get some time.

Thanks. I hope you found it useful.

If you mean to represent a load that the amplifier drives, you can yes. This would then need to be taken into account in its parameter calculations.

Don't forget the PNP, emitter to collector this time. Think of it as the NPN that I show, reversed. 🙂

No worries. I hope it's useful.

Hi, the STM32CubeIDE will automatically create most of the code for you (with the integrated CubeMX). If you pause the video you can then copy the code I add to write to the LCD. For the image header file, I create this using a program called Bin2c which is freely available from Segger. It can read a binary file (in this case a 800*480 BMP file) and outputs a C array, which you can then use to form/copy into the test_image_splash.h file. The tool is available here www.segger.com/free-utilities/bin2c/ I use a BMP format that is 16 bit, 656 as a source.

I'm glad it helped.

No problem. Hope it helped.

Hi, I create this using a program called Bin2c which is freely available from Segger. It can read a binary file (in this case a BMP file) and outputs a C array, which you can then use to form/copy into the test_image_splash.h file. The tool is available here www.segger.com/free-utilities/bin2c/ I use a BMP format that is 16 bit, 656 as a source.

No problem, I'm glad it helped. It's probably due to the use of Eclipse (I think) so one less thing for STM to worry about! The biggest issue for many are the #include dependencies....they can be a nightmare when you first get started.

Hi! Looking at the User Guide, the earlier boards were shipped with no demonstration firmware. So, if you have one of those, then maybe not. You may be connecting like this already, but CN16 is used for USB programming (ST-LINK), not the CN15 USB. Also, check powering by looking at the user guide (UM2033 Rev 5) page 18. A default demo that you can flash with the ST Programmer is available in the Tools/Software..Binary Resources here : www.st.com/en/evaluation-tools/32f769idiscovery.html# Hope that helps. If not, I think if you follow my tutorial (provided you have a good board) you should be able to bring it to life.

@@stevenfenton3319 Thanks you Man ! you were right. I had to flash the demo .

Hi, good question. The Re in this case enables the circuit to maintain the Vbe required. Without it, or if it's chosen as a small value, the transistor has the potential to begin to cut off during the negative AC input cycles. You can get rid of it if you're only interested in the +ve cycle of your AC input, thus saving a resistor. You might see this done, say in a class B arrangement. You can choose +ve or -ve depending on your choice of npn/pnp. In the video I demonstrate the effect of choosing different Re values. Generally, i usually choose a value of Re that gives me a Ve of approx 10% of the supply (as a rule of thumb) but this is purey down to the application really. It also helps to reduce the amplifier gain (-rc/re) offering another form of feedback. I hope that helps.

@@stevenfenton3319thank you for the explanation that was so clear , I also found some configuration of the self biasing amp that has voltage devider for biasing instead of only Rb and i didn't understand the advatages and disadvatages of each one

@@mohameddrissi1075 if you use potential divider bias, this helps to fix the DC base voltage independently of the Beta. The Vbb (base voltages and currents) are based on the potential divider. The expense is another resistor, so it really depends on your application.

@@stevenfenton3319 thank you sir , that helped me so much But right know i'm really struggling with how to design class ab amplifier all text books / videos / forums fixes values for resistors without explaining why or claculating it , no formula to calculate Icq could you please help me with that ?

@@mohameddrissi1075 That's on my 'todo' list.

No problem. I'm glad the video tutorial helped. Please subscribe to receive updates on the other STM32 tutorials I have in the pipeline.

You're welcome. I hope the video helped.

Yes, you are correct. This was part of the intro, but for completeness, you integrate over the period (not forgetting to rescale the expression based on that period), eg. -T/2->T/2, and scale 1/T.

They need to be copied from your FW support files. These can be downloaded using the IDE or by going to ST.com. They're usually stored in your FW repository.

@@stevenfenton3319 Thank you!

BSP will compile with errors until you add the needed modules. I think 2

@@johnbaker3745 Did you ever find a set of modules to add that seemed to solve this problem? I am getting compilation errors relating to BSP

You're welcome. Hope it was helpful.

Glad it helped.

Glad it helped.

Sorry, yes, I will GitHub it if I get a chance.

Indeed, if you convolve in the time domain, you multiply in the frequency domain. But, as I've shown if you multiply two functions in the time domain, this is equivalent to the convolution of their 'spectra' in the frequency domain. Note I change from time to frequency in the equation. I hope that clarifies it.

@@stevenfenton3319 Yes, it does. Thanks!

Thanks, I'm glad it was useful.

Glad it was useful.

No problem. I configured it as follows : #define I2S_WS 12 #define I2S_SD 17 #define I2S_SCK 23 So, all mapped to IO12, IO17 and IO23. I hope that helps you. Thanks for the interest.

Thanks.

Glad it helped.

Glad to hear that

Thankyou. I'm glad it was helpful.

Thank you, glad it was helpful.