Excessive hum

[Rebel]

Hi all, i have made a fender pro junior but when i connect the power transformer i have a big problem : the hum of 50Hz is very loud!! the circuit is not yet in its case and i'm testing it on the working table.
the scope shows that the cables of trasformer are full of 50Hz noise, i have shielded all i can but the noise is still here!
could you give me some suggestions?

[Chuck H]

50hz is going to be from the AC mains. It could be inducting from the PT into the OT due to proximity or orientation. Did you follow a layout? How are your transformers oriented?

[loudthud]

Are you sure it's 50Hz and not 100Hz? Did you install the 47 ohm resistors to ground on each side of the 6.3V heater supply?

[Rebel]

thank you for the rapid answer!

loudthud
you're right i didn't install the 47ohm resistor because on the schematic this kind of resistors are "Fire Proof" and in italy i can't find them.
yesterday i have read on a schematic the importance of these resistors an i used two normal 90 ohm 2 watt. Is too much? the noise is decreased by 90% but a little hum is still here.
tomorrow i will use 47ohm 2 watt
Can I ask you a question?
how this 47ohm resistors work? they make a low pass filter with the secondary winding of transformer or are used for balance the current or voltage between heaters?

Chuck H
on the working desk, Output and power transformers are distant about 30cm,the power transformer is on the top left of the desk in this position "--" and the output transformer is on the bottom right in this position " | "if i change the orientation of OT and Pt transformers i can't listen any change of noise.
i didn't use a layout, i have made a pcb by myself it is a problem?

tomorrow i will receive the box for my ampli,it is 40*20 cm with 1,5mm of iron.
I hope that further reduces the hum.

[loudthud]

you're right i didn't install the 47ohm resistor because on the schematic this kind of resistors are "Fire Proof" and in italy i can't find them.
yesterday i have read on a schematic the importance of these resistors an i used two normal 90 ohm 2 watt. Is too much? the noise is decreased by 90% but a little hum is still here.
tomorrow i will use 47ohm 2 watt
Can I ask you a question?
how this 47ohm resistors work? they make a low pass filter with the secondary winding of transformer or are used for balance the current or voltage between heaters?

Inside the transformer there is a small amount of capacitance between the high voltage winding (the red wires) and the heater supply (the green wires) because generally the heater supply is wound on top of the high voltage winding. Usually you can measure a substantial voltage (perhaps 100VAC or more) to ground with a DVM if the 47 ohm resistors are not installed. The 12AX7 tubes pickup a small amount voltage and amplify it. The 47 ohm resistors shunt this voltage to ground. They also balance the voltage on each side of the heater supply to ground. 90 ohms is ok, most commonly 100 ohms is used.

Chuck H
on the working desk, Output and power transformers are distant about 30cm,the power transformer is on the top left of the desk in this position "--" and the output transformer is on the bottom right in this position " | "if i change the orientation of OT and Pt transformers i can't listen any change of noise.
i didn't use a layout, i have made a pcb by myself it is a problem?

tomorrow i will receive the box for my ampli,it is 40*20 cm with 1,5mm of iron.
I hope that further reduces the hum.

Depending on how the ground is layed out in the power supply, you may get a 100Hz buzz. The most critical area is between CR1, CR2 and the ground side of C15 and C16.

[Rebel]

thanks for the answer, now i know a little more
Can i ask you another question?
I know that the 50Hz hum is from the power line but i don't know the 100Hz buzz.
What could cause this noise? could you explain it to me?
is the 100hz buzz the second harmonic of power line at 50hz?

[Chuck H]

The 100hz comes from the rectified voltage. The AC starts out looking like a sine wave. After the rectifier the DC looks like a series of humps side by side, or to put it another way, twice the positive swings and no negative swings. Twice the positive swings means twice the frequency. improper grounding and/or bad filters and/or filtering can allow the buzz to impress on the signal chain and get amplified like a signal before it is filtered.

[loudthud]

Rebel, you have a healthy appetite for knowlege of how things work! I suggest you pursue an education in electronics. Most formal training these days centers around digital electronics and things such as power supplys and analog circuits are not covered very well. But there are many older books you can find that cover the subject in detail.

[Rebel]

I hope it's a compliment
I'm at the third year of electronic engineering and electronic is my hobby, i have built some guitar pedals and now I decided to build an amplifier. But at the university you don't study the effect of 50Hz or the armonics of it in audio circuits. i have the "Radiotron Designer's Handbook" and i'm studing it in my spare time.
The problem is that 50Hz is hard to eliminate! and when someone gives me a solution i want to know the effect on the circuit and why it happens.

[loudthud]

It was a complement! Here at Ampage (music-electronics-forum) there are members of all experience levels. It's hard to know what experience level a member is at when they ask a question but sometimes the question itself gives a clue. The Radiotron Designer's Handbook is old, but one of the best books ever written on the subject. As I said above, todays electronics education does not spend much time on power supplys and analog circuits.

Most solid state oscilloscopes made in the last 40 years will measure time. Once you can display a steady waveform, you count the number of horizontal divisions of one cycle, multiply by the sweep speed that the horizontal time base is set to, and you obtain the time period for one cycle.

The time period (in seconds) for one cycle is the reciprocal of the frequency. 1/50Hz = .02 seconds commonly refered to as 20 milliseconds. Also note that if you know the time period of one cycle, the reciprocal of the time period is the frequency. 1/.02 = 50Hz. So you have to do a little math to get the answer. The time period for 100Hz is half the time period for 50Hz.

[Rebel]

thank you for the complement
At the university i learned to use oscilloscopefunction generator and other tools.
My problem is that i know the importance of 50Hz buzz but not the importance of its armonics, for example i didn't know 100Hz hum. So thank you again
Returning to the noise i had on C15 and C16 340-350Vdc and 15V of ripple.
Now I have added 47uF 450V capacitor on C15 and C16 and so the ripple on them is decreased at 3V.
At the output transformer i have a signal of 6Vac from the push-pull (at 30% of pots), the question is this : 3V of ripple is negligible compared to 340VDC or is loud compared to the signal?

[loudthud]

The 340-350VDC supply is referred to as "B+", pronounced Bee Plus. 15V of ripple is not considered excessive in a guitar amp. At low volumes you will not see it at the speaker terminals. The high plate impedance of the power tube plates prevents it from getting to the speaker. When you drive the amp hard enough to clip the output, some of the ripple gets through to the speaker but the sound is so loud you can barely hear the ripple.

R32, C17, R33 and C18 form a filter that removes almost all of the ripple from the B+ before it gets to the preamp. This is where any ripple on B+ will get into the signal path. If adjusting the Volume control changes the volume of the hum, that means that hum is getting into the signal path either through the B+, a bad ground, proximity of the signal to a transformer wire or possibly a bad tube.

Modeling programs are quite popular now and are used often in educational programs. SPICE is one such program. A simplified program that just models power supplies is popular with members of this forum. Download a free copy here: PSUD2

Below is a screen shot of a simple power supply. The green trace is the voltage coming out of the transformer. It is slightly distorted because current flows in little bursts. The yellow trace shows the little bursts of current in the transformer. The red trace is the B+ at the filter capacitor. Notice how there are two bumps in the ripple for every cycle coming from the transformer. This is because the special arrangement of the rectifier diodes, called a full wave bridge, connects the output of the transformer in the correct polarity to provide power on both halves of the sinewave. I set the frequency to 60Hz because that is the frequency of power in the USA. The ripple therefor is 120Hz. The program allows you to set the frequency to 50Hz.

It takes a little practice to use the program effectively. There is a forum for users on Duncan's website where you can get some help.

[Rebel]

thank you for the program i will use it as soon as possible.


This is the circuit i have built.
I have the problem that you mentioned in your post! when i increase the volume the hum increases too.
the four black capacitors on the top left are C15,C16,C17,C18 is possible that the first tube on the left (V1) get the 3v of ripple and amplificate it?
i can't solve my problem

[loudthud]

.is possible that the first tube on the left (V1) get the 3v of ripple and amplificate it?

As I said above, R32, C17, R33 and C18 form a filter that removes almost all of the ripple from B+ before it gets to the preamp.

There are several problems with the amplifier in the picture.

General principles:

There are wires that radiate noise and wires that are sensitive to noise pickup. Keep them as far apart as possible. Also, keep them as short as possible. Generally this means keep AC and transformer wires far away from any wires that carry the audio signals, especially the input jack and wires to and from the Volume and Tone pots. Wires to and from the Output Transformer are not very sensitive but they need to be kept away from the input and control wires.

Shielding is not needed when components are correctly layed out.

General layout of chassis components:

The power transformer is too close to the input jack. It should be located on the other end of the chassis. It can be close to the Output transformer but needs to be oriented for minimal hum pickup.

The Power Switch is in the opposite corner from where the power cord connector is located. This will require you to route noisey AC wires all the way across the chassis.

Printed Circuit Board layout:

AC wires for the heater supply go all the way across the board. This means that high currents are flowing through the preamp.

AC heater wires a snaking around in the preamp. This allows hum to enter the signal path when signal tracks are next to the heater tracks. Many commercial amps that use PCBs employ DC heater supplies in the preamp for this reason.

The grounds are poorly layed out. The ground for the filter caps, C15 and C16 should not be grounded near the preamp. Ground should flow from the power supply past the output stage connecting to grounds there, then the phase inverter (V2) and finally to V1 and the input jack. C17 should be located near the power tubes, C18 should be located near the preamp and grounded there.

[Rebel]

so the pcb is not correct
The problem is that this layout was made by a not qualificate man, and i have to use it for an examination on 30 september so and i can't change everything.
do you think that if i use twisted wires for heaters it will be better?
or do i have change the layout?