Hello. I've made a subsonic filter for my portable speaker by following the schematic in Elliott Sound Products website. (Project 99). I did my first ever pcb. So far so good. I used a single power supply op-amp (LM358) instead of dual. The problem is: when I hook it up to my portable speaker and use THE SAME power supply as the amplifiers use (no matter battery, laptop power brick), the speaker starts crackling at less than 1/3 of the volume (sounds like overly agressive clipping). But if I connect the filter to SEPERATE power brick no problem occurs, can play at full volume with zero distortion. I'm wondering does it actually need a seperate power supply and why, or can it be something, like power cables' length? I wish you could give me a workaround to somehow power it from the same power supply, as powering just subsonic filter alone from secondary psu (18650 for example) seems too complex and not worthwhile. Thank you, your help is greatly appreciated!
cof — imgbb.com
cof — imgbb.com
cof — imgbb.com
cof — imgbb.com
Last edited:
That circuit uses a bipolar supply. Can you post the exact schematic
that you used?
For one thing, you would need an output coupling capacitor
for a single supply circuit.
Last edited:
I followed the exact same schematic, just replaced op-amp with a single supply version. I guess that might be the problem?
Yes, there must be input DC biasing networks for the op amps
to set their DC outputs to the midpoint of the supply, as well as
an output coupling capacitor.
Connect two 10k resistors in series, across the power supply.
Connect a 100uF capacitor from their center to ground with + to the center.
Connect the ground ends of the two existing 220k resistors to the center
also, instead of to ground.
Connect a 10uF capacitor in series with the audio output, with + to the TLO72.
Make sure that the circuit ground is connected to the negative supply terminal.
Post your updated schematic here for verification, before altering the circuit.
Last edited:
Ok, I'll post the updated schematic here tomorrow, although I won't get all the modifications right at the first time, that's for sure. I'm still wondering though, why does it work flawlessly when powered from seperate psu as it is? Thanks for your help.
Right now the power supply floats wrt the audio ground.
The supply may be grounded through the amplifier when that is connected.
The outputs will not be DC centered properly regardless.
+1 @rayma for the suggestions.
Running from a single supply requires an offset (virtual gnd). Using a zener to stabilize that new virtual gnd (vcc/2) might be a further refinement.
IMO the OP should ask the moderator to move this thread to http://www.diyaudio.com/forums/analog-line-level/ and it might get some more comments as its really analog not digital line level.
Running from a single supply requires an offset (virtual gnd). Using a zener to stabilize that new virtual gnd (vcc/2) might be a further refinement.
IMO the OP should ask the moderator to move this thread to http://www.diyaudio.com/forums/analog-line-level/ and it might get some more comments as its really analog not digital line level.
Here are the original and modified circuits, as @Rayma explained to do it. The additional parts are named after their values for easier recognition, in case I made a mistake. I also drew a bright red line connecting the supply minus to circuit gnd, did you mean it the way I did it by saying: “Make sure that the circuit ground is connected to the negative supply terminal.“ ? Thank you a lot!
original — imgbb.com
modified — imgbb.com
original — imgbb.com
modified — imgbb.com
Last edited:
The modified — imgbb.com schematic looks correct with the red short connected as drawn.
C9 could be shifted to be in parallel with C8, the 100uF 1/2 B+ decoupler could be bypassed with a lower value solid cap (150nF PP), and the output cap 10uF could be replaced with a Bipolar/NP/PP type for slightly lower distortion.
All 150nf and 100nF caps could be replaced with these PP caps - Panasonic 150nF Polypropylene Capacitor.
Some might argue the cost/performance benefits of upgrading to PP but it is not a costly experiment and hey this is DIY.
Dan.
C9 could be shifted to be in parallel with C8, the 100uF 1/2 B+ decoupler could be bypassed with a lower value solid cap (150nF PP), and the output cap 10uF could be replaced with a Bipolar/NP/PP type for slightly lower distortion.
All 150nf and 100nF caps could be replaced with these PP caps - Panasonic 150nF Polypropylene Capacitor.
Some might argue the cost/performance benefits of upgrading to PP but it is not a costly experiment and hey this is DIY.
Dan.
Looks ok (including the red line). Of course, you don't need C9 since it is shorted.
Make sure the capacitor voltage ratings are larger than the power supply voltage.
Thank you both guys! I'm going to etch a new pcb tomorrow and keep my fingers crossed
. By making the suggested board I can connect it to the same battery as the amplifiers, right? As a side note, I'm going to have some spare time this summer so I may replace the passive crossovers with active. Can they raise the threshold at which tweeter's amplifier starts clipping or does an amplifier clip the most ONLY in low frequencies so the tweeter getting only high frequencies from passive xover is relieved anyways? I hope you get my question.
Xappys;5460201h said:
The HF can ride on top of the LF, so filtering before the amps can give more headroom
Last edited:
Use good quality machine pin IC sockets, you won't regret it.
110-44-308-41-001000 Mill-Max | Mouser
Use good quality pp film signal capacitors.
Also add a 100k resistor across both the audio input and output terminals.
Last edited:
I use a similar IC socket right now 😉 .
I'll use a crossover from the same page: Linkwitz-Riley Electronic Crossover . It's 4th order linkwitz-riley and I guess I'll still have some questions regarding single supply operation. Some material about why shall I add 100uF cap, 10K resistors why these values and where to place them would be greatly appreciated and will probably save you guys some hours of explaining me the same stuff all over again. 😱
Does C8 as in my schematic go to circuit gnd (marked 1) or supply minus (marked 2), after I remove the C9?
modified schematic (link to img)
modified schematic (link to img)
The C8 is connected directly across the supply voltage input, observing polarity.
The audio circuit ground and the power supply minus are connected directly together.
That is, connect together 1, 2, and the bottom end of C8. These are all "ground".
Last edited:
- Status
- Not open for further replies.