F5 - is normal ICs twist out the Hum?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello,

Thanks Jorge for your help, but I have some questions to confirm.

I made a little diagram of what you said, and I look like this:

The Amp GND of right and left (R1 and R2) and R10 will not connect GND PSU?

No, they will connect to the ground side of the RCAs, each to each channel.

About the big currents from the loudspeaker (from the minus output binding post) of course they must be connected to the 0 point between the capacitors ( or near it, like you have in the schematic)
 
Last edited:
....................The big currents working in loudspeakers should not return to the center point of the capacitors?...........
Yes.
And the big Charging Pulses around the rectifier circuit MUST NOT be allowed to contaminate any other grounding Wire/Cable/Plate/Trace.

This advice MUST NOT be followed. It is wrong!!!!
1- The wire from the loudspeaker return, the black output terminal, must be connected to the 0 of the power supply, near the electrolytic capacitors.
The Zero Volts between the Capacitors must never become the Audio Ground. Take a wire or trace from the Zero Volts connection to a remotely located Main Audio Ground. this link can be long or short or very short. It MUST be a link that DOES NOT pass charging pulses along the link.
 
Last edited:
Everything remains the same ... but I must tell you that F5 had a very difficult childbirth that I will describe, because it can be related to some of the effects I have described...

I had everything set up (had not connected the network will F5), but a person (friend) had already built an F5 gave me some advice and told me not to connect any mass, neither the speaker nor the Amps.

I saw did not respect the scheme but believed, turned it on but nothing was heard, but I tried to narrow down Bias was all unstable, and then began to hear a sound but distorted sound that could not be the F5.


I disconnected and then ask understood what to do now! He told me to pull the same termitores ace of mosfetes washers, and repeated not to connect because no mass lay with floating masses!!

I pressed the F5 and after a few seconds a thermistor burned like a match, hung everything and went back to talk to the person counselor.

A person came to my house, Bias values ​​were at 0.50V but sound nothing suddenly he turns on a mass of amps, and the voltage increased to 2V at R11 and R12, the fuse blew, it was hot, a burnt R11 such high currents.

We changed the pots to the minimum values​​, put new resistance R11 and new fuse, but as it was very hot the F5 did not give any sound, I think that some of the internal device Mosfetes blocked.

We waited a few minutes and yanked, this time connected with the masses, but the tension in R11 and R12 on the right side of the plate which had burst and the thermistor resistance could not tune the Bias, 2V and kept us was always off.

We removed the card from the place and saw that a lane next to the pot and R11 had melted, the others were good, with a shunt put the card ok, and then everything is back to normal.

I commented this summary because there may be some correlation between some of these facts and this little speaker from 20cm to hum ...


Connections of grounds, I have done everything, now I can only do very little:

- Get another source or other capacitors

- Other PCBs Amps

- Mosfets replace the board that took the scalding, or even JFETs?


I leave here a summary of the measurements made you an hour with a Bias of 1.3 A:

- DC voltage on the speaker load

** Left = 0.02 V Right= 0.069 V


- AC voltage in the rails (after the bridge)

*** Left = 0.25 V AC Right= 0.63 V AC


The tension in my home network is 226V AC.

The pre service is in Rotel RA-1250

Thanks for reading all this.

Blink
 

Attachments

  • Imgp5258.jpg
    Imgp5258.jpg
    209.4 KB · Views: 134
  • Imgp5253.jpg
    Imgp5253.jpg
    191.3 KB · Views: 131
  • Imgp5255.jpg
    Imgp5255.jpg
    236.3 KB · Views: 131
  • Imgp5269.jpg
    Imgp5269.jpg
    92.6 KB · Views: 134
Last edited:
Hi,
Attached it is an updated in the wiring using one of the PS board picture. This is the way I will wire it. One thing I noticed it is that the ground do not make start point. Cable to too long. So I did some changes in the way I will wire it. I think it compared with the powerflux wiring.
 

Attachments

  • PS board.jpg
    PS board.jpg
    218.6 KB · Views: 102
Why? Can you please elaborate? Thank you!

TS has now created a common impedance with the red wire going to power grond. all voltage that drops over that wire is shared, and thus even to the input ground.

VERY BAD indeed.
at least try my previous scheme (re-attached), or this Alternative option for your PSU board, with star ground.

i've attached your original too.
 

Attachments

  • f5hum7.jpg
    f5hum7.jpg
    72.7 KB · Views: 150
  • f5hum73.jpg
    f5hum73.jpg
    68.9 KB · Views: 158
  • f5hum72.jpg
    f5hum72.jpg
    64.3 KB · Views: 148
Last edited:
Take a wire or trace from the Zero Volts connection to a remotely located Main Audio Ground. this link can be long or short or very short.

You are wrong!!!! This link MUST be very short, because the currents from BOTH speakers will pass thought it and will develop a spurious voltage.

Is a common mistake to think about Ground like only the 0 Volts reference, actually the power supply capacitors are were the AC currents close. Ideally we must have a "short" for AC between Positive, Negative and Zero Volts Ground.

The place where this "short", for AC, take place is the main filter capacitor, so the link from the capacitors 0 volts to the 0 volts ground signal reference, will add to this AC impedance, and the link voltage drops will add to the power supply AC voltage. The input stage is very sensitive to this AC spurious voltage, because the "input" of the output MOSFETS are referenced to the positive and negative rail, in the configuration used (not the more usual follower configuration).

So, please keep the link the shorter as possible...
 
Last edited:
VERY BAD indeed.
at least try my previous scheme (re-attached), or this Alternative option for your PSU board, with star ground.

i've attached your original too.

Hi Powerflux

A AMP unhappily don't work without a source, so, think for a moment in your first option, the one you say is a better alternative. In the moment you connect a source, the commom ground of the source will shunt both RCAs grounds together, even worst, they do it remotely at the end of interconnect cables, and you end with a nice and BIG ground loop.

We must think in the concept of grounding, in real operation conditions... ;)
 
Hi Powerflux

A AMP unhappily don't work without a source, so, think for a moment in your first option, the one you say is a better alternative. In the moment you connect a source, the commom ground of the source will shunt both RCAs grounds together, even worst, they do it remotely at the end of interconnect cables, and you end with a nice and BIG ground loop.

We must think in the concept of grounding, in real operation conditions... ;)

you are right :)
 
Hi, Blink_PT

But you still have the hum? Even without interconnects, means, without nothing connected to the input RCAs?

I ask because i really don't understand it, from your last post.


Hello Jorge,

I already said above, I have no hum without interconnets.

Simply connect an IC that I hum in the opposite column, then call the two ICs and got such a hum in the two columns.

I've used all ICs I have, and they are many, shielded, unshielded, with only the side of the screen Amp, with only the screen side of the source. Everything, everything has been tried on ICs, since Arataca (space MIL 27500) to the most expensive.

The problem is that the F5, but where?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.