Help, hum in an equalizer!!!
all frq have hum
18 0 18v 750ma
7815 & 7915 with heat sink
2200uf 25v caps
100nf 100v caps
1 amp diode
how to solve it???
please help!!
all frq have hum
18 0 18v 750ma
7815 & 7915 with heat sink
2200uf 25v caps
100nf 100v caps
1 amp diode
how to solve it???
please help!!
1. Faraday's Law: Changing electromagnetic (EM) fields in the air will induce changing currents in any conductive loop. And changing currents in any conductive loop will induce changing EM fields in the air. (Both are in proportion to the geometric area enclosed by the loop, all else being equal.)
That means that any space between the wires in a pair will create an ANTENNA.
Both receiving and transmitting WILL happen if you have any open area between two wires that form a loop (i.e. "a circuit").
One easy way to get hum is to have "receiving antennas" feeding opamp inputs (or maybe even "receiving antennas" at outputs, before some other amplifier or system).
Then all you need is AC Mains in the air.
First check if it's OK when metal case is closed up, especially if you have fluorescent tube lights there.
Next, check your AC input wires and any transformer-to-rectifier and rectifier-to-caps wiring, to see if they are working as transmitting antennas. If they are not tightly twisted together, all the way to each end, that is bad and they are probably transmitting well. See below.
First, eliminate the typically-worst "transmitter" antennas: Tightly twist together any AC input wires, ALL the way to each end. They should never be apart. Also, do the same for the wire pair(s) from transformer to rectifiers, and from recifiers to smoothing caps. If PCB traces were used for any of those, they should stay extremely close together, everywhere.
Next, eliminate the typically-worst "receiver" antennas: Signal and Signal Ground pairs!
Always keep each signal and its signal ground extremely close together.
That means either have them in a shielded twisted pair (shield is NOT signal ground, then, and should be connected to chassis ground but only at ONE end), or, a tightly-twisted unshielded twisted pair (5 or 6 turns per inch, ALL the way to each end).
If you use PCB traces for signal and signal ground, then the signal and signal ground traces should stay extremely close to each other, everywhere. There should never be a large gap between them. Using a signal-ground plane in the signal input and signal processing areas of the PCB can make it easier to do layout. (Also, do not mix signal ground with other grounds! See farther below.)
You should also twist together all DC power/gnd pairs, and output pairs.
2. A changing current in a conductor will induce a changing voltage across the conductor's parasitic inductance and resistance.
So ANOTHER way to get hum is to let the reference ground for an opamp input resistor share any length of ground-return conductor with ANY other type of ground-return, on their way back to the power supply.
If you do that, then at the non-power-supply end of the ground conductor, there will be a changing voltage, NOT zero volts.
If that changing voltage is connected to the "ground" end of an opamp's input resistor, then that changing voltage will be arithmetically summed with the opamp's signal input voltage. That would be "a BAD thing".
That is why "star grounding" was invented.
At the least, have a completely separate ground-return conductor for the signal input ground reference points for the opamp inputs, all the way back to the single "star ground" point in the power supply. Do not use that ground return for power decoupling capacitors, or anything else except the input resistor grounds.
If you already tied them all together with the decoupling cap grounds, on the PCBs, maybe you can hope it's not the problem, unless everything else has been eliminated.
P.S.
Each board should probably have at least one (or more) large-ish capacitor for each rail, maybe 220 uF or more. But that is probably not closely-related to your problem. Also, on each opamp power pin, I usually used 10 uF and 0.1 uF in parallel, to ground, but that is probably not related to your problem and it looks like you might already have them on there.
Cheers,
Tom
Last edited:
very good!
It is just thumbnail size.
Please upload full size schematic.
Thankyou.
Thank you ashokavarthanan sir!
Also please share component side layout (silk screen)?
If possible can you export copper side pcb layout to a4 pdf format? because I need darker print for toner transfer method to make pcbs.
Thank you .. regards..
Also please share component side layout (silk screen)?
If possible can you export copper side pcb layout to a4 pdf format? because I need darker print for toner transfer method to make pcbs.
Thank you .. regards..
- Status
- Not open for further replies.