Drop resistor value

[Painkiller82]

Hi,

I received all my parts for a LM3886 build today, and I'm starting to build now.

I have a 2*22V transformer, which will be about 31V after the rectifier. I have read that 28V i max for this chip when I use 4 Ohm speakers. What will be a good resistor value to drop the voltage to 28V?

Espen

[audioDIYer]

Don't...

You have a 2 x 22V transfo...

Before the rectifiers, you have 22 x 1,414 = 31.11V
After the rectifiers you have 31.11V - (2 x 0.6V) = 29.91V

This is ok to drive LM3886 into 4Ohm IF you are using TA package...

Look here

[Painkiller82]

I'm using this schematic: http://diyaudioprojects.com/Chip/LM3.../LM3886_CA.htm

[edit]: I'm using the TF, the isolated ones.

[Painkiller82]

Can I run the TF on 29.9V, or will it burn? Will it be a problem unless I run it on max volume?

[richie00boy]

It will be unreliable and may fail taking your speakers with it.

You can't drop with a resistor as that will not drop a constant voltage with variable load and will wreck the impedance of your power supply. You need a voltage regulator or a new transformer. Probably cheaper to get a new transformer.

[audioDIYer]

With TF ones, it will be a little hot.
If you have big heatsinks with good ventilation, there is no problem, even into 4 Ohm load.

I have 2 LM3886TA on a 10 x 10 x 3 cm heatsink driving 2 x 4 Ohms load (stereo). It goes hot, but not too hot... Only 45ºC
The transformer is a 2 x 22V 260VA

I think you will not have any problem if you listen music at moderate levels...

[AndrewT]

Originally posted by audioDIYer
Don't...

You have a 2 x 22V transfo...

Before the rectifiers, you have 22 x 1,414 = 31.11V
After the rectifiers you have 31.11V - (2 x 0.6V) = 29.91V

This is ok to drive LM3886 into 4Ohm IF you are using TA package...

Look here

you forgot to take account of transformer regulation which will make the chip run at higher quiescent dissipation.
This on it's own will not kill the chip but will require a surprisingly large heatsink. That combined with the TF package and this chip will run hot.
These all conspire such that the Spike protection will turn on sooner on transient peaks.

[audioDIYer]

Originally posted by richie00boy
It will be unreliable and may fail taking your speakers with it.

LM3886 have internal protection. I have mine working for 3 years...

Originally posted by richie00boy
You can't drop with a resistor as that will not drop a constant voltage with variable load and will wreck the impedance of your power supply. You need a voltage regulator or a new transformer. Probably cheaper to get a new transformer.

LM338T
http://myweb.tiscali.co.uk/nuukspot/...ub.reg.psu.png

or
LT1084
http://www.linear.com/pc/productDeta...40,C1055,P1281

You choose...

[Painkiller82]

Thanks for the replies. I will try with what I have for now. My speakers says 4-8 Ohms, but I measured the DC impedance as 3.9 Ohms.

Page 2 (Nut) shows my speakers. http://www.mamut.net/dynabel/isis_geb_nut_set_brosj.pdf

Can you tell me anything from that?

[SheldonD]

Originally posted by Painkiller82
Thanks for the replies. I will try with what I have for now. My speakers says 4-8 Ohms, but I measured the DC impedance as 3.9 Ohms.

Page 2 (Nut) shows my speakers. http://www.mamut.net/dynabel/isis_geb_nut_set_brosj.pdf

Can you tell me anything from that?

Yes I would say you are very safe.

I have speakers that are DC resistance 4.1Ohms but dynamic impedance stated 8 Ohm, this is almost identical to your case.

The DC resistance is not the impedance seen dynamically with music which is AC, it has to be considerably higher as there is considerable inductance in voice coils and crossover .

My chipamp 3386tf the insulated one.

In my case with 22 ACVolt 400VA supply playing very loud organ music the heatsinks i/2 by 5 by12inches got warm to about 105 deg F

I also ran with regulated using LM338 knocking Voltage DC from ~30V to 25 V (you must have a dropout to make the regulators work). This to me worked even better as I could increase the caps at the chip which improved the bass tremendously, by preventing voltage sag under load.