Forums :: NEW! Web Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Oct Wed 21, 2020 12:52 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 15 posts ] 
Author Message
 Post subject: Ideal envelope shape for soft mute
PostPosted: Oct Tue 06, 2020 9:58 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
I want to optimize a soft muting circuit. Soft mute removes the clicks from the sudden beginning and ending of an audio waveform. I made one using the charge and discharge of a capacitor and connected the + of the cap to the VCA volume control pin of an audio amp IC. It works on the charge side (attack) , but the discharge (release) drops too suddenly and produces an audible thump. If there was a simple way to make it a mirror image of the attack, that would work well.

So I've looked into soft mute circuits found on the web.All the soft mute circuits I've tried have some level of thump either at the attack or release or both. Without going fully digital, I'm thinking of building a soft mute circuit with a programmable envelope. It would be a 4017 IC configured as count-to-N and halt. Clocked at 100 Hz, it would take 0.1 seconds for the ten outputs to cycle. Each output would have a different value resistor, which, in series with a common resistor to ground would make up as stepped voltage divider that feeds the VCA volume pin of the audio amplifer IC.

There may be other ways to skin this cat, and I'm open those with minimum parts count and parts that will be available for decades.

So the question is, what is the ideal shape of the envelope? Fast rise at the begging and slower at the end, as in the image below? Or would a straight line ramp be better? Certainly not a mirror image of the discharge (release) in the following image?
THANKS in advance for any light shed here.


Attachments:
capchargedischarge1.jpg
capchargedischarge1.jpg [ 40.6 KiB | Viewed 338 times ]
Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Tue 06, 2020 11:50 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2394
Location: Saskatoon
When I was playing with a morse code generator program, I used an S-shaped function for attack and decay.
Attachment:
Dahdit_600Hz_48ms.png
Dahdit_600Hz_48ms.png [ 212.45 KiB | Viewed 321 times ]

If you're getting a thump, that tends to be a sign that there's a DC bias on the audio that's also getting switched on and off.


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Wed 07, 2020 3:29 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
HI, Bob, I can see how that envelope would make a soft attack and decay. I believe I remember hearing it in one of your postings.

BobWeaver wrote:
If you're getting a thump, that tends to be a sign that there's a DC bias on the audio that's also getting switched on and off.

I'll check for DC, I don't see how it could be present. The amp is a TDA1013. The volume level is set by a voltage controlled gain pin that produces silence at 2 V and full volume at 7 V DC on the pin. It's hard to imagine that Phillips would design a DC bias into the chip that would be affected by the gain control pin.

Also, the audio output is capacitor coupled. The thump occurs only on the fast dropping voltage in that wave form. It seems that if there was DC involved, it would be what was ramped up at the attack so there would be a thump there, too.

Finally, when I connect the wiper of a pot set as a voltage divider to the gain pin and quickly flip it on and off, there's no thump. Not very elegant, but to make my point, I could connect a solenoid to a pot to quickly flip the sound on and off.


Last edited by Macrohenry on Oct Thu 08, 2020 2:39 am, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 08, 2020 2:08 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2394
Location: Saskatoon
Where would a person find a datasheet for the TDA1030? The only one I could find online was from Wuxi Youda Electronics, and there was no information on the voltage controlled gain (or much else for that matter).


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 08, 2020 2:39 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
Whoops, that's a typo. I've corrected it in the previous posting. It's a TDA1013. Datasheets are indeed on the web, I've attached mine to this posting.

Following up on previous posting, I measured the voltage at the speaker when the thump occurs. There's only AC voltage, no DC. So the thump is coming from somewhere else. I suspect the rapidly discharging voltage on the capacitor.


Attachments:
TDA1013B.pdf [109.77 KiB]
Downloaded 6 times
Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 08, 2020 3:49 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2394
Location: Saskatoon
When I think of a thump (as opposed to a click), it reminds me of some of the older cheap single ended audio amps that would send a pulse of current to the speaker on power up, as the big output electrolytic charged up to half the supply voltage. How fast does your voltmeter respond? Would it pick up a pulse like that. Have you tried it with the voltmeter connected to pin 2 of the IC to see if there is a level shift there?
It seems strange to me that there would a thump on the decay but not the attack.

You could try the following circuit. The two transistors work as constant current source and sink, to charge and discharge the capacitor at a fixed rate (until the cap gets near the supply limits where voltage tapers off). The output is pretty close to the 2-7 volt range.
Attachment:
EnvelopeGenerator3.png
EnvelopeGenerator3.png [ 10.81 KiB | Viewed 254 times ]


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 08, 2020 4:59 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
Well, thank you! This is about as simple as I'm looking for. I'd already tried constant current sources to give a straight ramp. BTW, you recommended a CCS when I was characterizing Push Pull transformers for single ended use. I studied up on that and then used a LM317 configured as a CCS and it provided clear and uniform results.

I've also seen a perfect ramp wave on my scope by using a CCS to charge a cap. However,I've not been able to achieve modeling that in LTSpice using the LM317 model or a dedicated CCS chip. There's always some curve to the line.

Same is true in the case of this circuit you provided. Below is the LTSpice model, which was much quicker for me to cobble up than the real thing. Two items of note: First, the output is backwards from what I expected. I can work around that, and I don't know why I was expecting the pulse to charge rather than discharge the cap. Second, not that it matters, but according to LTSpice, a five volt pulse doesn't match the top the pulse, as shown in the second simulation graph.

You prescribed 1uF and I ran the simulation at 10uF as it gives a lot closer waveform to what I want. YMMV. Electronics is amazing.

Anyway, this circuit meets my design goals of always readily available parts and simplicity, and the rise and fall are close to symmetrical, PLUS is goes from the prescribed 2-7 volts. How did you come up with it?


Attachments:
image2.jpg
image2.jpg [ 152.67 KiB | Viewed 250 times ]
Image3.jpg
Image3.jpg [ 90.72 KiB | Viewed 250 times ]
Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 08, 2020 5:37 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2394
Location: Saskatoon
I've used those one transistor constant current sources quite a bit. So, linear cap charging was the first idea that popped into my head. I just stuck a source (top transistor) and a sink (bottom transistor) together. They don't look very linear in this application, because the voltage gets a bit too close to the supply rails, but that shouldn't hurt in this case.

You can make the attack and decay time asymmetrical by changing the values of either or both emitter resistors. You can also switch the input over a different range. For example you can switch between 0 and 10 instead of 0 and 5. That will also affect symmetry.

You can also adjust the minimum and maximum output voltages by replacing the emitter resistors with voltage dividers that go from the positive supply to ground with the emitter connected to the tap point.

Edited to add:

Macrohenry wrote:
I've also seen a perfect ramp wave on my scope by using a CCS to charge a cap. However,I've not been able to achieve modeling that in LTSpice using the LM317 model or a dedicated CCS chip. There's always some curve to the line.


I've built some very linear sawtooth generators using these one transistor circuits. The trick is not to get too close to the supply rails, and don't use ceramic capacitors. This is one application where you can really see the non linearity of ceramic caps. I don't know why LTspice would have a problem.


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 08, 2020 6:10 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
Good stuff. Forgot to answer your question about voltmeter observation. It's a Fluke digital and all you can see is a rapidly changing display. With a scope set to slow almost crawling trace, the ramp up and down is visible. Best I can tell it's at my desired 0.1 second rise time. The rise matches the LTSpice simulation pretty much on the spot. Not so much on the decay.
Which is why I started thinking about the 4017 approach.

I dug out a couple of CDS optoisolators. They have a slow response time, but still too fast to be used alone. When used in combination with a simple parallel resistor and capacitor to ground, using just one completely eliminates attack and decay thumps. Maybe it's not so much of a problem, but CDS cells are heading toward unobtanium, which is why I like your circuit.


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Mon 12, 2020 10:58 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
I've found the sound I want. Soft attack and decay. Here are scope traces of attack and decay playing a staccato note. I have to look hard to see remaining fluorescence as the trace decays. Note how asymmetrically odd the envelope is shaped. This is because of the CDS LDR that's in parallel with 40 uF and 8.2K, all connected from the IC VCA pin and to ground.

I can't seem to post an animated gif, so here's a link:

https://www.dropbox.com/s/m48ih5r63z2nx ... n.gif?dl=0

These photo are composites of four and 5 photos respectively:
Attachment:
composite.jpg
composite.jpg [ 104.21 KiB | Viewed 186 times ]


Attachments:
LDRVCA2.jpg
LDRVCA2.jpg [ 81.1 KiB | Viewed 184 times ]
Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 15, 2020 6:07 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2394
Location: Saskatoon
I couldn't view the gif. My computer says the file is corrupted.

What are the part numbers for your LDR optocouplers? I don't think these parts were ever very common.
I believe there's still a Canadian company that makes an expensive high end version that has fairly fast response (if you need fast response).


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Thu 15, 2020 10:05 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
BobWeaver wrote:
I couldn't view the gif. My computer says the file is corrupted.

What are the part numbers for your LDR optocouplers? I don't think these parts were ever very common.
I believe there's still a Canadian company that makes an expensive high end version that has fairly fast response (if you need fast response).


It's an animation consisting of the five gifs you can see in one of the composites. You're not missing much.

I prefer slow response. These two were from goldmine-elec.com part number (Silonex or Advanced Photonix) NSL 7053. I can't find a datasheet but there is discussion on forums that say that's probably a part number for the epoxy container. Same forums say the specs of the goldmine LDRs vary wildly. I've used VTL 5C3 and 5C4 in the past, which were about as consistent quality you can get, but they just can't compete with digital methods if you need precision.

I used the Goldmine in my auto volume circuit, and it works fine from unit to unit. In this application I tried a bunch of varying units with no appreciable difference in performance. Pots are needed anyway to set parameters. http://www.tompolk.com/hobbies/automati ... r/avc.html


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Fri 16, 2020 1:59 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
I fixed the animation. I had forgotten to optimize it before saving. It works when I open it in Firefox and turn off my animation blocker. As stated above, you're not missing much. https://www.dropbox.com/s/m48ih5r63z2nx ... n.gif?dl=0


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Fri 16, 2020 7:04 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2394
Location: Saskatoon
Yes it's working now. It does help in visualizing what's happening.


Top
 Profile  
 
 Post subject: Re: Ideal envelope shape for soft mute
PostPosted: Oct Fri 16, 2020 11:25 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4135
Now that the animated gif is fixed, I no longer get the error message from ARF. Posted below:


Attachments:
LDRVCAanimation.gif
LDRVCAanimation.gif [ 909.2 KiB | Viewed 67 times ]
Top
 Profile  
 
Post New Topic Post Reply  [ 15 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: Google [Bot] and 8 guests



Search for:
Jump to:  


































-->


Privacy Policy :: Powered by phpBB