Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Hum with single-to-split power supply for the equalizer.

Status
Not open for further replies.

BrunoARG

Full Member level 4
Joined
Aug 15, 2014
Messages
214
Helped
38
Reputation
76
Reaction score
38
Trophy points
28
Location
Buenos Aires, Argentina
Activity points
2,177
Hum with single-to-split power supply for equalizer.

Hello people.

About a year ago I built an audio amplifier which sounds very very well, but it has been continuously making a pretty annoying 100Hz hum that I couldn't ever fix.

It consists of two parts: A 4 channel mono equalizer, and a 4 channel (actually using only 3 but equalizer channels are summed up to get a flat response) output stage made with TDA7294 and a passive filter for each one, and sounds very very loud and its bass response is excellent (with a 10" woofer).

The power supply comes from a single 12V transformer I got from a desk lamp (60W). The output voltage is about 16 or 15V (depends of the AC line) and for signal proccessing (equalizer and preamplification) I splitted it in lower symmethrical voltages (to scape from that annoying ripple, which can make it down to 13 or 12V at maximum volume.).

The problem I have is, when I connect my phone directly to the power stage, the amplifier sounds great, no hum and so loud. But when I connect it to the equalizer and then it to the power stage, there comes the hum, even with no music playing. The circuit is the following and the simulations show it should work fine.

Supply.jpg
Circuit diagram

Supply_SIM.jpg
Split power supply voltages

Supply_SIM_SIGNALS.jpg
Amplifier 1KHz signal from preamplifier and entering power amp input stage (seems well).


Do you know if there is a trick or something I didn't take into account? I will make some measurements in the actual circuit to show, when I can.

Thank you in advance.
 

Re: Hum with single-to-split power supply for equalizer.

First of all, I wouldn't say that any amplifier with "a pretty annoying hum" is working "very, very well"; it's not. You need to find the source of your hum. It's either coming from a poorly filtered dc supply or bad shielding on your input or bad pc layout.
 

Re: Hum with single-to-split power supply for equalizer.

I'd look for a "ground loop" or ground offset kind of
problem. Battery powered phone has no choice but
to follow the shield ground but a mains powered EQ
may have significant ground "signal". Splitting a
supply with a center tapped xfmr could add to this
if phase loads are asymmetric and case != ground
plane. Check also the possibility of ground-stud
(back panel) connections, as used to be seen on
turntables and other low level input devices, to
connect case-ground to case-ground outside of
the audio connection electrical path.
 

Re: Hum with single-to-split power supply for equalizer.

My entertainment system made an annoying hum when I hooked it up to my laptop computer, while laptop power was from mains. The hum stopped when I ran the laptop on batteries only.

I believe the problem was a ground loop, because the hum stopped when I removed the incoming cable tv. In fact I only needed to unscrew the plug so as to disconnect the ground connection only. Even though it was an inconvenience, I was glad to find a solution to getting rid of the hum.
 

Re: Hum with single-to-split power supply for equalizer.

Thank you for your fast responses.

The transformer is not connected to ground, even the plug has not the third earth pin, so there is not a ground loop here..... I will try to connect the circuit GND to earth to see if it gets better. As there is not any that long connection between boards, I don't think that will solve the problem, but it worth trying. When I have some time, I will measure the equalizer output respect from 0V (actual circuit ground) to see where comes the noise from.
 

Re: Hum with single-to-split power supply for equalizer.

Hello BrunoARG,
Are you using mains earth in your project?
If so, then you need to connect the body of the transformer to ground
(earth) as well.
It also wouldn't hurt to insert a 100uF electro and a 0.1uF MKT cap
between VSS and ground.
Regards,
Relayer
 

Re: Hum with single-to-split power supply for equalizer.

100 Hz hum is usually generated by poor power supply filtering (frequency doubling of full wave rectifier), not capacitive or inductive coupled mains voltage.

The schematic in post #1 doesn't completely clarify all possible paths of 100 Hz interference, but feeding an insufficiently filtered supply into the audio path via R6 is basically a bad idea. Also the filter effect of the simple discrete voltage stabilizer shouldn't be overestimated.

Did you ever connect a stable power supply instead of the scarcely filtered input voltage for test?
 

Re: Hum with single-to-split power supply for equalizer.

Just because your transformer is not connected to ground doesn't mean you can't have a ground loop. Maybe if it was called a "return loop" or a "common loop" that would make more sense.
 

Re: Hum with single-to-split power supply for equalizer.

Well, finally I got some time with my amplifier close to the oscilloscope so I said "hey it's a good time to measure its signals and supply" so I did.

I found out the problem: First, the output signal (from equalizer) was a VCC/2 with some ripple... strange.

I measured the output of the filtered supply and it was about 10V (I think it's because of overloading the transformer). The regulator which is actually a LM317 at 7V got that 10V with ripple supply, and I think that's the problem. Well, actually it is, the output of the regulator is a ripple signal "clipping" at 7.1V, so I am supplying the equalizer with basically noise, which goes to the filters and to the output.

I did not realize to save the waveforms to show you, but it's clear to understand.

No ground loop problems or EMI, just a bad power supply here. I hope this can help anyone.
 

Re: Hum with single-to-split power supply for equalizer.

Do you really need a two transistor splitter to find half supply? It would probably work just as well with two equal resistors and give better supply tracking.

Brian.
 

Re: Hum with single-to-split power supply for equalizer.

Well, actually it wasn't so necessary, but as I didn't have time to test the 2 resistor virtual ground, I just put the transistors to assure low impedance, which I think is good to avoid noise in the supply. But it works pretty well, and it's more efficient than two resistors, but more complex without any doubt. As equalizer drains about 25mA, it would be 250mV drop on a 100R resistor, or almost 50mV on a 22R. 7V / 22R is more than 300mA just in splitting the supply
 

Re: Hum with single-to-split power supply for equalizer.

The zener diode/emitter follower has no error amplifier so its voltage regulation is pretty bad and produces ripple.
The two transistors voltage splitter is feeding half of that ripple directly to the (+) input of the opamp with no filtering.
The 2 transistors splitter can produce maybe 100mA but its load is the (+) input of the opamp that might use a current of only 1uA (LM833) to nothing for a modern audio opamp with Jfet inputs. A high resistor values divider with a filter capacitor or two of them will work much better at removing the ripple. Replace the zener diode/emitter follower with a real voltage regulator IC.
 

Re: Hum with single-to-split power supply for equalizer.

For a circuit with signal ground referenced to negative supply, you basically don't want a symmetrical rail splitter because it feeds half the power supply ripple to your virtual ground node. Instead, the virtual ground must be strongly bypassed to the signal ground respectively negative supply.

In multi stage amplifier, there's always a risk to generate an unwanted feedback path through the virtual ground if it doesn't expose sufficient low impedance. It's often better to use individual bypassed voltage dividers for each amplifier stage.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top