 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : low pass filter
Page 1 of 3   |
|||||
| Author | Message | ||||
stanleyella Guru  Joined: 25/06/2022 Location: United KingdomPosts: 2134 |
|
||||
| stanleyella Guru Joined: 25/06/2022 Location: United KingdomPosts: 2134 |
oops. My question -- is the vga sound low pass circuit only for vga or can it be used with non vgaa pico mite. It's simpler than the non vga low pass circuit, no chokes.. just ordered 10 and just one preferred value cap.  |
||||
thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 4047 |
It can be used for either. I believe that once updated both manuals will show the same filter with the inductor/choke as that is a better quality filter. Best wishes, Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6812 |
The filter shown above will work, but the output shouldn't be fed into a modern amp with good speakers. You are ok with 32R headphones or most cheap desktop speakers (probably). It has pretty poor performance. The 1k / 47nF combination extracts the audio from the PWM. The other two resistors merely halve the output level. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
Turbo46 Guru Joined: 24/12/2017 Location: AustraliaPosts: 1611 |
Due to RC filters in the input and in the feedback circuits most modern amplifiers that I have seen usually have an upper frequency roll off beginning at a bit over 20kHz (not all though). That would add another two stages of RC filter to the one shown above. I honestly don't believe that much 44kHz would reach the tweeters in a modern HiFi system. My calculations show that the corner frequency of that filter above is around 3.4kHz. For most applications I believe that is adequate. Having said that my father used to say that 'if a job is worth doing it's worth doing well'. For that reason I applaud the efforts of Volhout and matherp in trying to design a more ideal filter.  It should make those with 'golden ears' happy.I also believe in 'horses for courses' and, as I personally cannot hear much beyond 5kHz, anything more than that is just a waste of effort. There are sites on the web you can use to test the frequency response of your hearing. Try it, you may be surprised. Bill Keep safe. Live long and prosper. |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6812 |
There's not a lot of point, from an audio perspective, in being 3dB down at less than about 10kHz. Not many adults can hear fundamentals above that even if they think they can. However there are harmonics there that add additional "air" to the sound, particularly on percussive instruments such as triangles and cymbals. This isn't about what you can hear though, it's about what you can't hear. I was surprised to find that many modern amps are pretty flat up to about 100kHz. They don't have to be expensive to achieve that. Even the LM4871has only dropped from a gain of 50dB at 10kHz to 40dB at 60kHz. That 40kHz is a real sound and is really driving the output devices in the amp, and heating them up as they shove unwanted power into the load. Just because no-one can hear it doesn't mean that it isn't there. It's like the RF oscillations that can burn big MOSFETs out if you don't put a gate resistor on them, but not quite as extreme. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Turbo46 Guru Joined: 24/12/2017 Location: AustraliaPosts: 1611 |
I would be happy with a two stage RC filter at around 5kHz. If anyone wants to roll their own RC low pass filter I have made a small spreadsheet that will help.  Starting with about 220R (about the min. load resistor that you would want to use) a capacitor to give you the frequency that you want. Adjust the resistor if needed. Add a second resistor in series with the original. Make it 10 times the value of the first one so that there will be little interaction. Then add a second capacitor to 0v 10 times smaller than the first one. Finish up with a resistor divider to give you the output voltage you desire. Done. I found that 330R with 0.1uF followed by 3300R and 0.01uF will give me what I want. 4.8kHz. If anyone wants a copy of the spreadsheet I can upload it here. It's in LibreOffice .ods format but I believe Excel can cope with that. Bill Keep safe. Live long and prosper. |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 3816 |
How much of the 40kHz / 44kHz is still there after that? John |
||||
| Turbo46 Guru Joined: 24/12/2017 Location: AustraliaPosts: 1611 |
I’m sorry John, I don’t know and I don’t really care. Somebody who does care may be able to help if you really need to know. Bill Keep safe. Live long and prosper. |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4254 |
 Residual 44kHz is 38dB down, so roughly 50mVpp Note that your calculation for a single stage RC filter gives you -3dB at 4.6kHz. Chaining 2 stages will be -6dB at 4.6kHz (also shown in above graph). Numbers say nothing. When you are happy with the result, that is what counts. Volhout Edited 2023-03-27 20:26 by Volhout PicomiteVGA PETSCII ROBOTS |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 3816 |
Thanks. John |
||||
Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1115 |
To be on the safe side, I connected an additional 10uF Capacitor in series to the headphone/Line output to eliminate the DC component. Edited 2023-03-27 21:46 by Martin H. 'no comment |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6812 |
If you are feeding an amplifier you could probably go down to 2.2uF rather than 10. If you want headphone output then 100uF would be much better. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| stanleyella Guru Joined: 25/06/2022 Location: United KingdomPosts: 2134 |
The remaining 44kHz not filtered is still a signal that needs amplifying. This can overload the amp and then the amp adds inaudible harmonic distortion. I've had tweeters blow. At our age we remember before computers, people wasted their money on "hi-fi". OK there were "white" goods and 3 piece suites but that's different. Remember lava lamps. I remember building speakers in the 90's with speaker box designer but that needed Theile and Small parameters, resonant freq, Q and Vas but most speakers don't give this info and it's hard to measure yourself. I still got a stereo,2 wharfdale diamond IV, Sansui amp.. with "super feed forward" connected to the tv for input. There's a 2TB usb hd connected to the tv and that's got all my music on it in mp3. Not hi-fi but sounds loud and clear and bassy. |
||||
| Turbo46 Guru Joined: 24/12/2017 Location: AustraliaPosts: 1611 |
So roughly 20mV RMS. Seriously? 20mV at 44.1kHz can do all that? Bill Keep safe. Live long and prosper. |
||||
| Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1115 |
44Kilohertz is the sample rate, i.e. the carrier frequency of the audio signal. The usable audio signal can only have a maximum of half the sample rate To smooth Signal, the Lowpass should be at around 18 to 20 Kilohertz (half the sample rate). In this range are frequencies that only children, cats and dogs can hear. 'no comment |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4254 |
The issue with 44kHz is you can't hear it, and the amplitude is always the same. So when you play audio from a WAV file, or, to lesser extend, play sound at low volume, you may be tepted to increase the volume of a connected hifi system. And that could damage tweeters. 44khz is the carrier, the sound is the modulation. Volhout PicomiteVGA PETSCII ROBOTS |
||||
| ville56 Regular Member  Joined: 08/06/2022 Location: AustriaPosts: 96 |
Would it be a possibility to add a RC notch filter centered at 44 kHz before the lowpass filter? Probably there would be the need for "mild" adjustment of the notch frequency. There are a some RC circuits around on the web, but i don't know which one can be made adjustible easily. What is the experts opinion on that? Reagrds, Gerald 73 de OE1HGA, Gerald |
||||
| stanleyella Guru Joined: 25/06/2022 Location: United KingdomPosts: 2134 |
I just built the low pass filter and there's a 1 per second clicking with this code for c=50 to 1000 play tone c,1050-c,5 pause 5 next c Edited 2023-03-29 03:14 by stanleyella |
||||
| Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1115 |
Gerald, here you find some useful background informations: low pass filter a pwm signal into an analog voltage in the example they use a 100kHz PWM carrier frequenz. So they create at first a 50Khz Lowpass. Should be nor big deal to recalculate the L R and C values for our 44kHz carrier. Edited 2023-03-29 05:02 by Martin H. 'no comment |
||||
| Page 1 of 3 |
|||||
 Print this page |
