DX Blame MkIII - 2013 builders thread.

Jump to Latest
#121
Terry,

how old are those main caps .....?
Click to expand...

I'm not sure. At least 7 years old. I bought them before I got out of the hobby. They have been in a drawer for at least that long. I did reform them when I took them out. I just measured and I see about 24mVAC across the rails so maybe that is part of the problem. I will substitute in some known good caps and see if that helps.

Thanks, Terry
 
Last edited:
#122
BV said:
Common PSU is conected to two amp modules, and input cables shields are than conected again together.
Click to expand...
I'm not sure what you mean by input cables shields connected again. The shield of my input cables are the ground wire for the input. it only connects to the RCA and the - input lug on the PCB. They become connected when the source is connected because the source plugs have a common ground. This is very common.
Ground loop, sensible to induction form magnetic fields. Transformer somewhere nearby. Result is induced current, causing voltage drop at cables shielding resistance, this is known issue.
Click to expand...
I have the transformer as far from the input and output as I can.
Minimize loop area (twist input cables together, place power modules as close as possible) and minimize currents in input ground loop (use correctly 10R resistor separating power ground and input ground).
Click to expand...
You lost me here. What power module? I did install the 10R resistors. No change.

Thanks, Terry
 
#123
They become connected when the source is connected because the source plugs have a common ground. This is very common.
Click to expand...
Exactly, and hum in such topology (common PSU, unbalanced inputs) is also very common because ground loop induced currents (voltage drop at nonzero impedance of cables shields is added to input signal). Power module = one channel of amplifier..
If You insert 10R in such loop (in each channel one independent resistor), shield resistance and 10R form voltage divider for such disturbing signal, and hum is much lower. To one end of 10R can be connected only input bias resistor ground end, input cable ground (shield) and ground point of NFB divider, nothing more! Other end of this resistor is connected to power ground (center point of PSU).
It can help a lot, but it is not the best possible solution (dual mono, independent PSU, or true balanced inputs..)
 
#124
I
BV said:
Exactly, and hum in such topology (common PSU, unbalanced inputs) is also very common because ground loop induced currents (voltage drop at nonzero impedance of cables shields is added to input signal). Power module = one channel of amplifier..
If You insert 10R in such loop (in each channel one independent resistor), shield resistance and 10R form voltage divider for such disturbing signal, and hum is much lower. To one end of 10R can be connected only input bias resistor ground end, input cable ground (shield) and ground point of NFB divider, nothing more! Other end of this resistor is connected to power ground (center point of PSU).
It can help a lot, but it is not the best possible solution (dual mono, independent PSU, or true balanced inputs..)
Click to expand...

I can't understand your directions. Maybe something is being lost in the translation. I have built 14 amps. All with common PSU. This is the only one with this behavior.

Thanks, Terry
 
#126
Terry, look at this.. You must break this loop (to reduce induced ground currents) with inserting 10R between power GND and input GND, on each channel PCB (interrupting PCB ground traces, some "modifications" needed)..
 

Attachments

  • loop.jpg
    loop.jpg
    116.8 KB · Views: 388
#128
If posted pictures (schematic and PCB) are valid, than here are following faults:
To input signal ground should be connected only R01, C04, C06, C07 and one end of R31, 10R. To this point also should be connected input RCA ground , but input conector must be isolated from chasis. And nothing more!
To input signal ground can not be connected C15a, C15b,C11b C11a , it is AC short circuit from suplly voltage on signal ground (AC short circuit for 10R), via PSU main capacitors it is AC shorted to power GND! Also reconnect C01, R17, R08b to power ground.
In current state R31, 10R, have no function..
 
Last edited:
#129
OK, now I see. R30 actually lifts the grounds of everything connected to that trace. Probably not the intention. That is probably why Carlos suggested to use a jumper there. I will try a 10R inline with the input ground. I wonder why no one else has reported a problem with this? According to Carlos, there have been many of these boards built. I wonder is any of this batch has been built yet other than mine. Other than the hum this amp really sounds quite nice.

Thanks, Terry
 
#130
OK Guys, thanks to the kind help from all of you I believe I got to the bottom of it. The first thing I did was pull R30 and reinstall the jumper per Carlos' suggestion. Then I installed a 10R 1W resistor between the ground wire and RCA connectors and connected the shield to the case per BV's suggestion. That completely eliminated the added hum when attaching both sides from the source. I still had some noticeable hum so I disconnected the power supply and connected up my DX PSU with a 40-0-40vac transformer. The amp is dead quiet now. Looks like I will have to fork out for some new filter caps. 🙁
At least I know what it all was now. I would recommend installing those resistors at the RCAs. It is easy to do and really made a difference.

Blessings, Terry
 

Attachments

  • Input grounding.jpg
    Input grounding.jpg
    225.4 KB · Views: 357
#132
Hey I got rid of all the hum. I wanted to give the big transformer and caps another try. I still had the bad bum so I hooked up the DX supply once more and this time I had a hum. I noticed that the power cable was still attached so I unplugged it. The moment I did the hum went away. So now armed with this, I pulled the ground plate off my filter caps and enlarged the hole in the center so I could insulate it from touching the hold down bolt. I reattached everything using new mounting locations and now the amp is void of hum. I guess I will have to go with a bridge to attach the ground to earth. Whew, what a ride. Now I can finally concentrate with building some cases for the other 4 amps in need of one. 😀

Hopefully my journey will help someone else with their build.

Blessings, Terry
 
#133
Then I installed a 10R 1W resistor between the ground wire and RCA connectors and connected the shield to the case per BV's suggestion.
Click to expand...
Wrong, I never sugested that.. You only got lower level of hum, but for "blameless" function You need to change grounding (change PCB..), on upper side of 10R can be conected only R01, C04, C06, C07 and input cable shield. On the RCA conector the shield could be conected direct, and RCA (isolated from case) could by conected to case via 10R.
 
#134
Sorry BV, didn't mean to misquote you. Regardless, adding the 10r resistors at the RCAs solved the problem and insulating the center tap from the case cured the rest. The amp is dead quiet now. I will not be modifying the PCB. This amp is finally playing nice so I am on to other things.

Thanks, Terry
 
#138
I had to make one more change. Every thing was fine as long as something was plugged into the input but if nothing was plugged in, there was a bad hum. I had to short the input ground to the case to get rid of it. So my fix was to install another 10R resistor between one of the input grounds and the shield which is connected to the case. I'm not sure if a resistor is the best choice for this but it worked. If one of you thinks something else would work better here I am all ears. 😉

Blessings, Terry
 
#140
Hi guys are interested in the construction of the blame right but I got lost in many post and my English is not that great. I wanted to know if it was done the comparison with other amplifiers such as krell ksa50 and what it came out as a result. of course I mean the sound quality in addition to raw power. or if you can tell me the link where you see the comparison with others. thank you
 
You must log in or register to reply here.
Top Bottom