
Home  Forums  Rules  Articles  The diyAudio Store  Gallery  Blogs  Register  Donations  FAQ  Calendar  Search  Today's Posts  Mark Forums Read  Search 
Chip Amps Amplifiers based on integrated circuits 

Please consider donating to help us continue to serve you.
Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving 

Thread Tools  Search this Thread 
9th May 2012, 11:34 AM  #211  
diyAudio Member
Join Date: Jun 2010

Quote:
I would calculate the primary fuse like this: Code:
80VA = 230V * xA I = 0.35A Now there are three types, fast, medium, and slow. I know fast won't work because of the inrush current. But I have no idea whether medium or slow is the right choice. Quote:
Quote:
Quote:
cheers! hurtz P.s. will comment on the multichannel setup later! 

9th May 2012, 11:40 AM  #212  
Mark Kravchenko
diyAudio Member

Quote:
My best pop story is setting up a supply board at 3:00 AM. Plugging it directly into the mains socket and hearing a strange high pitch sound. And then boom, like a shot gun. All the caps blew their tops off. I had soldered them in backwards! So yes there can be noise in them there cans! 

9th May 2012, 11:58 AM  #213 
diyAudio Member
Join Date: Jul 2004
Location: Scottish Borders

The transformer can supply low current when the demand is low.
It can supply high current when the demand is high. It can also survive massive overloading without any damage if the transient duration is short enough. All these different loading situations can be imposed by a correctly operating amplifier. I find the best way to determine the lowest fuse to provide that house burning down protection for the secondary side is to use the amplifier rating as a guide to the fuse size. My rule and Quasi's rule are almost identical. Amplifier maximum output power leads to maximum output current into your nominal load. eg. 100W into 8ohms speaker is equivalent to 40Vpk and 5Apk into a nominal 8r0 load. The secondary fuse in each supply rail should be ~ half that value i.e. F2.5A. Quasi uses rms output current as the guide. The same 100W into 8ohms amplifier has 28.2Vac and 3.5Aac. He suggests a secondary fuse for each supply rail of ~F1.75A, i.e use F2A Both these rules have been stated in many previous posts.
__________________
regards Andrew T. Sent from my desktop computer using a keyboard 
9th May 2012, 07:49 PM  #214 
Did it Himself
diyAudio Member

Forgive me, I see your point now.
__________________
www.readresearch.co.uk my website for UK diy audio people  designs, PCBs, modules and more. 
9th May 2012, 08:39 PM  #215  
diyAudio Member
Join Date: Jun 2010

Quote:
Code:
U * I = P 40 * 5 = 200 (18V secondaries) 18 * 1.41 = 25.5V (And not 51V because 25.5V is the maximum voltage difference one can reach from 0V) Into 8r0 load: 3.2A But! Isn't this the maximum current one rail can draw at a time? That is one 3.2A fuse per rail?  Multi channel setup: I would set the Frontleft as master 0V and then return all the other 0V lines to this one, i.e. speaker return, lowlevel, and RCA. On the other AMP boards the power0V will of course also be returned to the frontleft. cheers! hurtz 

10th May 2012, 10:07 AM  #216 
diyAudio Member
Join Date: Jul 2004
Location: Scottish Borders

Power = voltage times current.
That rule applies to constant DC supplies and constant DC current to the load. If the supply varies or the load current varies then the rule becomes: Power = root mean squared of voltage times root mean squared of current, normally written as P = Vrms * Irms If you have peak current of a sinusoid waveform and peak voltage of that same sinusoid then the rule must be reworded to convert the peak current and peak voltage of the sinusoid back to root mean squared to give the same heating effect as the original rule. If using peak values of sinusoid waveform then the rule becomes: Power = Peak voltage times peak current divided by 2, i.e. P = Vpk * Ipk / 2 Remember this all goes back to equivalency to the heating effect of constant currents through constant loads, i.e. constant DC operation.
__________________
regards Andrew T. Sent from my desktop computer using a keyboard 
10th May 2012, 10:09 AM  #217  
diyAudio Member
Join Date: Jul 2004
Location: Scottish Borders

Quote:
__________________
regards Andrew T. Sent from my desktop computer using a keyboard 

10th May 2012, 11:18 AM  #218 
diyAudio Member
Join Date: Jun 2010

Ah! Thanks for the explanation!
So we conclude: 8r0 = 1.6A 6r0 = 2.2A 4r0 = 3.2A cheers! hurtz 
10th May 2012, 12:09 PM  #219 
diyAudio Member
Join Date: Jul 2004
Location: Scottish Borders

F1.6A, F2A or F2.5A and F3.15A are the available fuse values.
But, you have it wrong. 68W into 8r0 is ~33Vpk and ~4Apk. giving an F2A fuse for 8ohms speaker. 68W into 6r0 is ~28.6Vpk and ~4.7Apk. Giving an F2.5A fuse for 6ohms speaker. 68W into 4r0 is ~23.3Vpk and ~5.8Apk, Giving an F3.15A fuse for 4ohms speaker. Note that the 4ohms to 8ohms current ratio is sqrt(2), not 2. Quasi's version of the rule reduces all the fuse values by sqrt(2). This takes account of the rms value of a sinewave rather than the peak value that I use.
__________________
regards Andrew T. Sent from my desktop computer using a keyboard 
10th May 2012, 04:33 PM  #220 
diyAudio Member
Join Date: Jun 2010

But my peak Voltage is always the same! I get your calculation using 68W und then multiplying by sqrt(2) to get the peak values. What I don't get is why I should assume P and R as fixed values to calculate I, rather than R and U.
Since my guess would be that for short pulses "Spike" doesn't kick in and the maximum 68W rating doesn't apply! cheers! hurtz Last edited by hurtz; 10th May 2012 at 04:39 PM. 
Thread Tools  Search this Thread 


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
lm3886 single sided pcb  ryan750  Chip Amps  56  27th May 2015 02:35 AM 
Single sided lm3886 board, opinions?  Adam Eng  Chip Amps  5  26th March 2014 02:18 PM 
Lm3886 single sided my first try  ampimp  Chip Amps  0  14th October 2010 10:08 AM 
Single sided non inverting LM3886 PCB  tlaAudio  Chip Amps  35  1st April 2009 07:16 AM 
New To Site?  Need Help? 