Lm4780

[Cheng]

Hi all , Brian/peter,

I have juz finish wif the BrianGT LM4780.





but still facing some hum problem, even at zero volume.

The chips runs very hot on the heatsink after 1hr(is it normal).

Can I replace the Rm(mute) wif a SMT resistor??

Can I use a 1/4 watt resistor for the feedback resistor??


Cheers.

[DonoMan]

but still facing some hum problem, even at zero volume.

Try moving the transformer wires to face away from the circuit. Try shielding the transformer if needed. Check all wiring.

The chips runs very hot on the heatsink after 1hr(is it normal).

Depends how hot. If it's too hot to touch then use a bigger heatsink or turn down the supply voltage (which will lower your output, of course)

Can I use a 1/4 watt resistor for the feedback resistor??

P = V²/R. Use your your rail voltage RMS (if +/- 30V, for example, then VRMS is 30 / sqrt(2) for about 21.2V), square it, divide by the resistor value. Use a slightly higher power resistor. (I can't give you a straight yes/no because I don't know the size of the resistor or what your supply voltage is)

[Cheng]

Hi DonoMan,

Thanks for the input. By the way when I switch the source from DVD player to a CD player, the heat sink temperature did reduce and most importantly, the hiss & hum noise disappear!
Can someone pls enlighten me on this.

I will try to on it for longer period tonight to see if the temperature maintain.

Re: the trans / vin, I'm using a 250VA 25-0-25 center tap transformer. rectified to around +/-38V rail.

So Brian/Peter, may I have my question answered??

Thanks.

[runebivrin]

Judging from your pictures, you should torn the heat sinks 90 degrees. The way they're oriented now stops air from flowing between the heat sink fins by means of convection. But they still look quite small, and once you put the lid on, the air in the box will heat up quite quickly since it cant escape the box.

Rune

[digi01]

Originally posted by Cheng

but still facing some hum problem, even at zero volume.

The chips runs very hot on the heatsink after 1hr(is it normal).

Can I replace the Rm(mute) wif a SMT resistor??

Can I use a 1/4 watt resistor for the feedback resistor??


Cheers. [/B]

Try shielding the signal cable or twine each other.try connecting the CHG of each board to chassis ground.

use big heatsink or reduce voltage.

for Rm,any 10K resistor will do.my amp is totally quiet when turn on.

for the feedback resistor,I use 1/4w dale.it is ok.

ZANG

[Russ White]

Did you isolate you LM4780s from the heatsink? If so what did you you use? If not, you will likely have all sorts of problems.

[jackinnj]

Originally posted by DonoMan

Depends how hot. If it's too hot to touch then use a bigger heatsink or turn down the supply voltage (which will lower your output, of course)

P = V²/R. Use your your rail voltage RMS (if +/- 30V, for example, then VRMS is 30 / sqrt(2) for about 21.2V), square it, divide by the resistor value. Use a slightly higher power resistor. (I can't give you a straight yes/no because I don't know the size of the resistor or what your supply voltage is)

With a +/- 30V Rails and an 8 ohm load the correct heat sink should have a thermal impedance of 1.76 C/W -- yours are too small and they are enclosed in a metal cabinet so they're not getting any meaningful air movement. EDIT: Your heat sinks appear to be something like 6" long, 4" base width, 1" fins -- the C/W is ~ 3.7 in free air, but much higher in an enclosed space. If you can figure a way of getting a 12VDC fan to blow 100 LFM a minute across the heatsinks you willl get down to 1.7 C/W.

measure the output voltage, not the rail voltage, to determine power. Most DVM's will be good enough to a few hundred HZ at least. you can decrease the rail voltage

with respect to the hum problem -- first, use insulated RCA connectors for the inputs -- if they aren't insulated from the chassis that ground connection becomes part of a loop. secondly, try a shield around the transformer made of spare printed circuit material.

[Cheng]

Hi all,

Here is the update.

Disconnect from the DVD source & connect to a CD player, the hum/hiss disappear even volume up. erm.. is does the source output influences on the amp output noise?? also it does reduce the temperature abit(not confirm). But i though the metal casing it self does help to dissipate the heat?

I'm using a center tap 25-0-25 trans. Think i'm not going to step down the voltage( as I wan it to stay at it original vtg/design),
but will see if i can get a 12v fan or change to a bigger heat sink.

I'll get a SMT 10k resitor for the Rm tomorrow as it will be a more sensible way to solder into the board. (hopefully it will not blown.)
the feedback resistor i'll also change to a 1/4 watt, as the 1/2 watt i have are actuallly too big to be solder in.(hopefully it will not blown.)

Ya. I got a chip blown initially. replace it and now the chip were isolated from the heat sink using a mica washer of slightly a bigger size with only one hole for the screw and silicon (white cream) for the transfer of heat. i did not manage to get the heat sink wif two fitted hole ( i did not drill)

Mean while still troubleshooting on the tube buffer. will keep pluggin in for further verification of temp/sound or other problem. ( eventually i will shift it to another casing as tis is a test casing for diy experiment purposes.)

Thanks all for the inputs and suggestion. Will update again on the sound.

Enjoy the music

[Peter Daniel]

Sorry for not replying earlier, but I missed that thread somehow.

The amp does not have input coupling cap, and gain setting resistor (680R) is connected directly to the ground. This may be the reason why your DVD source does not play correctly. You might try coupling cap with that source, anything between 2 -4 uF should be fine.

As to the other resistors, you should be fine with 1/4W and SMD .

[jackinnj]

here's how I cool the bridged LM4780 -- the PCB has been removed --
I milled some aluminum stock to attach the fan to the heatsink. The heat sinks were tapped for 6-32nd screws -- nylon -- which means that the heat sink, with its connection to the negative rail (no mica washer, or insulating hip washers -- will not come in contact with the chasis.