Help!! TORI buffer buffer wont work

[yongchao]

I am making a Tori Buffer now. but after soldered everything, i connect it to my amp. it wont work on my small testing 4-ohm speaker. a lot "priak priak" sound, very annoying. I am using IRF510 , LM317 , 162ohm for Gate and 10ohm for LM317.. what happen... i m using a regulated power supply..

thanks for help..

 

[yongchao]

pls help, this simple buffer is very easy and useful. only required a few of components. pls help...................
and the output DC off set is in volt range. strange... i used dual supply.

 

[KTpG]

I am pretty sure that the Tori buffer requires a capacitor on the output between the MOSFET source and the load. Since it is SE... I would really assume so, especially since you are getting such high offset. Put a 470-1000uF cap on the output and try it again.

 

[yongchao]

Thanks a lot for your reply, KTpG..
I have tried using a 4700uF Electrolytic Cap on the output, there are no difference.
Also, i have enquire someone in the forum sometime ago about this buffer, he siad since the buffer required +/- power supply, so there should be no DC off set.

Anyway, i just fry my 10ohm resistor.i used 1/4 watt resistor on the prototype. i think should use 5 watt instead. wat do you think??

Besides, anyone tried this buffer before?? how is its performance??

 

[Mpemba Effect]

I've not built a tori amp but I have built a szekeres which looks very similar. Performance is excellent - you can see some pics here Do you not get any output at all? Have you whipped out the multimeter on it yet? Any chance of posting some pics, may be you've made an error or two.

 

[yongchao]

i think i sholud be feed back loop problem. thanks. i will go back to try...

 

[peranders]

First I think you should measure the current through the mosfet. Does this LM317 thing work? The wiring is OK?

 

[peranders]

Delete this please....

 

[Porksoda]

Yeah, definitely sounds like a problem with your LM317. Properly set up, that 10 ohm resistor should only take about .1 watts, well below 1/4 watt. If it's burning up, then it isn't connected properly. The current through your mosfet should be a .125 amps plus the current your load is drawing.

 

[peranders]

When the LM317 is working properly you should have 1.25 volts between pin ADJ and OUT.

 

[yongchao]

Thanks guys....
after putting feedback loop.. it works fine..

since the IRF510 is capable to supply current up to a few ampere, and the limitation is probably the LM317.. right?? so can we put 2 LM317 in parallel to bring up the limits.

How can we know how much current can the LM317 supply?? and the bias voltage (determine by the resistor value)?

can we put a few similar buffer in parallel?? (with low value resistor at its output.)

can we bias a IRF9510 with LM337???

Thanks a lot...

 

[yongchao]

actually i forgot to look at the schematic at the top of the webpage..
(it's all Korean i dun understand)

 

[yongchao]

bump..

 

[Porksoda]

As I understand it, the purpose of the Class A buffer is that the MOSFET is in its most linear region. I do not believe increasing the current beyond what a single LM317 can handle will get you much gain.

As for the maximum current the LM can carry, you might want to check the data sheet. Keep this in mind though: Power dissipation is voltage times current. So the voltage from input to output times the current through the LM is the power. If you look at the data sheet for your heatsink you will see a deltaT associated with a given power dissipation. If your ambient temperature range plus the deltaT puts you out of the LM's specified operating temperature range, you need to back off.

But checking the data sheet for current limits is your best bet.

Again, pure class A buffers are pretty hard core. I don't think you would get much out of paralelling them.

But that is just my opinion, and I am sure somebody out there will tell me I have no idea what I am talking about

 

[yongchao]

Thanks Porksoda...
but i need more information from you guys before i make a better version of
Mosfet Buffer...