Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Dec Sun 16, 2018 9:40 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 48 posts ]  Go to page Previous  1, 2, 3  Next
Author Message
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 16, 2018 2:28 pm 
Member

Joined: Apr Thu 12, 2007 3:24 am
Posts: 1721
Location: Milwaukee,WI
If your transmitter is going to remain stationary in a corner somewhere and not need to be portable, an alternative to expensive outboard modules is available on Craigslist or at local Music stores. They sell small light rack mount processors like Compressor/Limiters and Equalizers. Many used units as cheap as 30 bucks each.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 16, 2018 2:34 pm 
Member
User avatar

Joined: Feb Sun 02, 2014 9:13 pm
Posts: 1866
Location: Roanoke, VA
richfair wrote:
It maintains good peak control but is audibly working when pushed hard.

That is always the case with single-band processing - when you push it hard it creates objectionable artifacts. That is why multiband processing was introduced in the late 1970s and has long been the standard in broadcast processing.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
http://plymouthcolony.net/starcityeng/index.html


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 16, 2018 2:43 pm 
Member
User avatar

Joined: Feb Sun 02, 2014 9:13 pm
Posts: 1866
Location: Roanoke, VA
richfair wrote:
Years ago NRSC suggested a pre-emphasis equalization curve that was intended to overcome the reduced bandwidth of “typical” am receivers.

I find NRSC preemphasis objectionable when listening to music on my better vintage radios - it sounds too bright, but I can switch it out and listen to straight multi-band processing without the preemphasis.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
http://plymouthcolony.net/starcityeng/index.html


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 16, 2018 7:58 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 884
Location: Yukon, OK USA
Rich;

A couple of questions; shouldn't C4 be connected to ground? In all my other transmitters my antenna matching connects that way.

You omitted the LED used for tuning the antenna. While it was not a requirement it was useful. Is it possible to put it back in and then adjust voltage to the circuit for 9 volts and compensate for the drop across the LED? I have several other ways of tuning the antenna. The LED was just quick and already there.

I should have my parts in today or tomorrow and will get this circuit up on a protoboard and check it out.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 17, 2018 4:26 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 566
Location: Equinunk PA 18417
black85vette wrote:
A couple of questions; shouldn't C4 be connected to ground? In all my other transmitters my antenna matching connects that way.
Good grief yes! Elementary mistakes like this are embarrassing. I'll fix that for sure.
Quote:
You omitted the LED used for tuning the antenna. While it was not a requirement it was useful. Is it possible to put it back in and then adjust voltage to the circuit for 9 volts and compensate for the drop across the LED? I have several other ways of tuning the antenna. The LED was just quick and already there.
I did say "with or without the LED" in my original post because, since I thought this thread would be of interest primarily to those who tinker and homebrew it would be sort of assumed. But, I will put it back in because I agree that a tune-up LED is wickedly handy.

Quote:
I should have my parts in today or tomorrow and will get this circuit up on a protoboard and check it out.
Superb, I look forward to your results. Tonight I returned to this project with a variation that ran smack into more instability and frustration. My prototype boards are small, packed, with high impedance signals (fet gate) control that are haphazardly scattered 1/10th of an inch apart. This isn't rocket science and yet tonight it may as well be. There is something to be said for nailing things to a piece of wood.

In a former life I made a good living as an independent sound mixer for film and television. Dale, I am with you all the way on both of your points. There are strong arguments to be made against blanket pre-emphasis designed to "help" the sound reproduced by a gang of old radios with limited high frequency performance, at the expense of high quality sound or what is actually possible. I used the NRSC pre=emphasis guideline because it is an established and adopted work. That doesn't make it right or good, just makes it a point of reference. BTW, I did mention an optional pot instead of a fixed resistor to allow a range of adjustment. I was thinking homebrewer's would understand.

Regarding multi-band processing, Dale you are on point again. Multiband compression is a powerful tool, as are look-ahead limiters and phase-time spreaders (or whatever the proper terminology is, I think you know what I am talking about). I still am questioning my efforts in this thread. It would be much easier to look on Craigs list for a used Behringer compressor, or heck, a guitar pedal effect is probably an improvement over nothing. If I wasn't interested in chasing my own tale that is probably what I'd do!

_________________
-Richard


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Thu 25, 2018 3:12 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 566
Location: Equinunk PA 18417
I promised to post a revised schematic of an analog limiter processor for the LM386 transmitter using junk box parts. This is breadboarded and works as intended with one caveat. The limiter’s activity indicator causes audible low level pops on my breadboard. I suspect bad layout and poorly-routed ground currents from IC1B and its LED are the primary cause. The fet gate is very high impedance and sensitive to millivolt changes, not helped by the breaboard. Good pcb layout and attention to ground current paths will hopefully solve that problem. I plan to build this for real BUT UNTIL THEN I MAKE NO PROMISE. The indicator led could use some improvement anyway. For one thing, TL072 makes a lousy comparator, for another a simple on-off light is not very informative.

<images removed by richfair>
Designed for a single regulated 9 volt supply. Current draw is transmitter plus 16ma. My transmitter draws 17ma when the tune-up LED is bypassed (after antenna is peaked of course), so total draw is about 33ma. When I build for real I hope to use a single TL074, assuming layout is still workable.

Transmitter:
<Refer to top post of this thread if you want pre-emphasis on your LM386 modulator.>
Limiter:
<I'll continue this offline because I want to do this in the way I set out in my first post, analog and "homebrew". I do appreciate those who pointed out my drawing errors.>

_________________
-Richard


Last edited by richfair on Jan Wed 31, 2018 1:42 pm, edited 3 times in total.

Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Thu 25, 2018 8:57 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 566
Location: Equinunk PA 18417
<removed by richfair>

_________________
-Richard


Last edited by richfair on Jan Wed 31, 2018 1:34 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 30, 2018 4:53 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 884
Location: Yukon, OK USA
Macrohenry,

Since I also have a two tube transmitter I thought a stand alone compressor that could be used with any transmitter would be a good project. I built yours with a minor modification. The output level was too high for my transmitter and I had to run the pot too close to the ground end. Figuring that IC2A had some amount of gain I picked the audio directly off the wiper of R15 and eliminated the output amp stage. Now the pot is about at midpoint. Much better. Electronics Goldmine had the TL062CP which is the low power version of the 082. They work fine. I also did as you mentioned and substituted an LED for D3. It is nice to have an indicator of what the compressor is doing when setting it up. I found a nice spot that keeps the distortion at bay without overdoing the compression. It also keeps the Pandora commercials at the same level as the music. Since this is not integrated into the transmitter the bipolar power supply is not an issue. It is still on the protoboard for now. But the results are good so far. Thanks!

Rich;

If you want an LDR to experiment with I have 4 spare and would be happy to send you one. Just send a PM with a mailing address.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 30, 2018 6:32 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3660
black85vette wrote:
I built yours with a minor modification. The output level was too high for my transmitter and I had to run the pot too close to the ground end. Figuring that IC2A had some amount of gain I picked the audio directly off the wiper of R15 and eliminated the output amp stage.
Thanks for the feedback (pun intended.) It's always gratifying to know others have success with such a simple circuit. And it might be even simpler:

I wonder if you are lucky enough that your input levels are high enough so you can also eliminate IC1A. It would then be a single IC circuit. Since you are in breadboard, can you bridge or substitute the 100K R3 with a 10K resistor so that IC 1A is essentially unity gain? If it still works, I'd think you could eliminate one of the ICs.

BTW, R5 was originally a pot used during development, so R4 was there to ensure a minimum value. I see now that R4 can be eliminated even if you keep IC1A.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 30, 2018 7:32 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 884
Location: Yukon, OK USA
Macrohenry wrote:
I wonder if you are lucky enough that your input levels are high enough so you can also eliminate IC1A. It would then be a single IC circuit. Since you are in breadboard, can you bridge or substitute the 100K R3 with a 10K resistor so that IC 1A is essentially unity gain? If it still works, I'd think you could eliminate one of the ICs.

BTW, R5 was originally a pot used during development, so R4 was there to ensure a minimum value. I see now that R4 can be eliminated even if you keep IC1A.


My primary source is an iPhone and I have to run it at near Max to get the level up high enough to be compressed. So I don't think I can skip the Input Amp. Likely a device running at line level would be different. Easy enough to try it out and see.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 30, 2018 7:49 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3660
black85vette wrote:
Macrohenry wrote:
I wonder if you are lucky enough that your input levels are high enough so you can also eliminate IC1A. It would then be a single IC circuit. Since you are in breadboard, can you bridge or substitute the 100K R3 with a 10K resistor so that IC 1A is essentially unity gain? If it still works, I'd think you could eliminate one of the ICs.

BTW, R5 was originally a pot used during development, so R4 was there to ensure a minimum value. I see now that R4 can be eliminated even if you keep IC1A.


My primary source is an iPhone and I have to run it at near Max to get the level up high enough to be compressed. So I don't think I can skip the Input Amp. Likely a device running at line level would be different. Easy enough to try it out and see.


To that end, as an alternative, I wonder if your configuration would still work if you eliminated signal buffer IC1B? I can't recall how important buffering is at that point in the circuit. Easy enough to try, though.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Tue 30, 2018 10:59 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 884
Location: Yukon, OK USA
I eliminated IC2a and IC1a. To get enough signal I put the iPhone directly into C1 without R1/R4/R5. There is enough signal to work with but R9 is right at the end of its travel and somewhat touchy to adjust. I think I will replace it with a 10k pot and pad it to the right range to allow it to adjust over more of its travel and be easier to set. The good news is that the compressor appears to function fine with two of the four sections.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 31, 2018 3:36 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3660
Excellent! Love it. Thanks for the update.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 31, 2018 4:15 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 19342
Location: Warner Robins, GA
Richfair, there is a simple way to do it.

Get a Raspberry pi 3 (think that's the one) as Stereo Tools makes a version that will run on it.

You may then even be able to build the transmitter on the pi itself for a totally self contained unit aside from a keyboard and display.

Think the built in audio will work ok, but if not there is an audio card made for the pi that will work.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 31, 2018 1:15 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 566
Location: Equinunk PA 18417
Rick, thanks for your offer.
Tube, thanks for your comments, I have enjoyed many of your threads.

_________________
-Richard


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 31, 2018 1:45 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 19342
Location: Warner Robins, GA
You're welcome and thanks for the complement about my threads.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 31, 2018 3:17 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 884
Location: Yukon, OK USA
Rich;

Let me know when you finish your design. I agree with you that an an analog circuit integrated into the transmitter design is the optimum way to go. I have enough parts to build a couple more transmitters and would like one of them to be on a single PCB with the transmitter and compressor. Also have spare op-amps and FETs to build in the compressor. In playing with several compressors now I am convinced that the 386 / Oscillator combo Tx benefits a lot from audio processing on the front end. It is so sensitive to over modulation.

Thanks for your contributions to this little transmitter.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Jan Wed 31, 2018 4:26 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1429
Location: Mesquite NV 89027 (from Coventry, UK)
This is an interesting thread. Looking forward to the final design.


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Feb Thu 01, 2018 5:07 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 884
Location: Yukon, OK USA
Here is another approach. I had built one of these as a peak indicator to help setting audio levels. But I was thinking all it lacked was an LDR. So I hooked one up in place of the LED then just put the resistor part between pin 2 or 3 (which ever you used for audio input) and ground. Connected the input through a 10k pot and attached to pin 5 of the 386. You can just hook up to the audio circuit without having to having to insert or disconnect any components. Makes it easy to disable it if you want to. Oddly enough it actually worked. I used 2N3904 transistors mostly because I have about 200 of them. :-) C2 can be changed to approximate a release time adjustment. I verified operation with a dual trace scope to see at what setting it started to attenuate and to be sure the output signal held constant. For initial testing I used a regular LED so I could see what the circuit was doing with various input levels and be sure I had enough audio signal.

http://www.circuit-finder.com/categories/audio/sound-level-meter/741/peak-indicator

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Audio processing an LM386-ECS100 transmitter
PostPosted: Feb Thu 01, 2018 5:56 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 566
Location: Equinunk PA 18417
black85vette wrote:
Here is another approach. I had built one of these as a peak indicator to help setting audio levels.

http://www.circuit-finder.com/categories/audio/sound-level-meter/741/peak-indicator


Yes, this is one approach to an activity light and it is certainly better than nothing. It, like many to be found, detects only one polarity of audio signal swings. Audio peaks, though, can be positive or negative and usually not symmetrical. Having said that, this circuit can be set to turn on the LED when audio peaks are generally in the vicinity of true peaks and so may be good enough, especially for a part-15 transmitter! My original approach should give an indication any time the gain is being reduced, which can depend on attach/release time constants as well as peak levels. It is a subtle difference that may have little added value to the user. Your idea is a good and clever one that gets the job done.

The LDR circuit Macrohenry posted is one example where non-symetrical detection works pretty well. The LDR simply cannot respond fast enough to control near-instantaneous audio peaks. It controls overall volume well but allows near-instantaneous peaks to pass right through. That is why it can sound really good, but does not prevent overmodulation. The LDR does a sort of "average" power integration, similar to a filament bulb that lights from the heat of average power,it doesn't care what the actual waveform is. Whether a detector is 1/2 wave or full wave, the shape of the detected average is practically the same most of the time. For true peak limiting though, which an LDR can never do by itself, a full-wave detector is mandatory or else one must rely on a symmetrical clipper or just ignore the distortion (since this is part-15 after all). My earlier circuit gave a pretty good fullwave detector and is capable of very fast peak detection and control. I think having an LDR in front of it would be an even-better approach.

Rick, I will PM you with hope that you can mail an LDR to me. I am on a temporary but austere budget for the next few months, so your generosity is appreciated.

_________________
-Richard


Top
 Profile  
 
Post New Topic Post Reply  [ 48 posts ]  Go to page Previous  1, 2, 3  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: 35Z5 and 3 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB