F5 power amplifier

1mW is ~-47dB ref maximum power of the F5 (50W into 4r0).
If one had noise at a level of -47dB one would be in tears after spending all that money.

Output offset is a "noise", if anyone will allow DC to be a noise. It certainly should not be there and certainly not considered acceptable.

If high efficiency speakers are being used with the F5 then it's even more important to minimise the output offset.

Andrew,

I if were to add a input capacitor to the F5 to prevent inadvertent DC to creep into the amp, is a 1 mfd, MKP a good choice, without compromising the bandwidth of the amp too much?

Thanks in advance.
 
Member
Joined 2009
Paid Member
Marra, did you follow the right proceed for bias and offset setting?
That is without the bulb.

Yes this is without the bulb connected. On the left channel P1 does not alter the reading across the source resistor. P2 I can adjust. I raised the voltage to 0.12V and the offset changed so I turned it back to its lowest setting. With the right channel connected the p/s voltage started to rise then the 1amp q/b fuse popped.
 
Member
Joined 2009
Paid Member
Soft Start Module for Mains Supply
Avondale Audio soft start module

Fully Automatic Timing Sequence - No more dimming lights, pops, clicks or blown fuses!

Operates in line fully automatically or manually with a low current switch.

Handles up to 1KW.

Includes an on-board mains current breaker / fuse, spark suppression and transient clamping.

Size: 97mm X 72mm Plus, no soldering
Soft Start PCB
Avondale Audio TPX pcb

Available in PCB only form for DIY audio projects.

Comes with list of required components and building guide.
 
500VA would normally use a T6A fuse to survive repeated start ups over the years.
With a soft start you can probably use an F2A if the resistor/s reduce the starting current sufficiently.

To use an F1A, I suspect you would need 3 CL60 in series and another 50r of Power Resistors to reduce the start up current and gradually bring it on line as the CL60s heat up.

With +-132mF you may want to consider slow charging. NTC are best for this, located in the secondary circuit and again bypassed when the CL60s have done their job.
 
Andrew,

I if were to add a input capacitor to the F5 to prevent inadvertent DC to creep into the amp, is a 1 mfd, MKP a good choice, without compromising the bandwidth of the amp too much?

Thanks in advance.

Just to clarify, I am asking if a 1uf mkp is a good capacitor value to block dc input to the f5. I am assuming 101k as the 'R' value of the input. I am not sure if 1/(2*pi*r*c) is the right way to calculate the high pass frequency for this setup.

TIA.
 
Yes this is without the bulb connected. On the left channel P1 does not alter the reading across the source resistor. P2 I can adjust. I raised the voltage to 0.12V and the offset changed so I turned it back to its lowest setting. With the right channel connected the p/s voltage started to rise then the 1amp q/b fuse popped.

Hi Marra, this is most likely your problem. You need to adjust both P1 and P2 - don't just turn one of them all the way to the end. Don't worry too much about DC before you have the current dialled in on the .47 ohms.
 
Member
Joined 2009
Paid Member
Thanks for the responses.
Andrew regarding the soft start I don't think theres any point in posting a link as there is no circuit diagram just pics of the pcb. I have been using one of these s/s in my Aleph 5 clone for the past 6yrs without any problem supplying 300VA transformers feeding +- 47,000uf per channel and a 2A slo/blo fuse.
Christian on one channel I adjusted P1 almost fully upwards without any voltage reading across the source resistor; P2 slowly raised the voltage upwards. The right channel when connected with the left disconnected blew the fuse. I was using a 1amp fuse which is what it says in the original article but having reread the article I realise it was advised in conjunction with a variac whereas I just turned the amp on. I guess I need to borrow a friends variac or try it with the reccomended 1.25amp slo/blo and see what happens.
Thanks again Marra
 
Official Court Jester
Joined 2003
Paid Member
you must iteratively turn one then other trimpot ;
while doing that - monitoring both - voltage across one source resistor and output offset .

while waking up one mosfet - other one is still sleeping , so current which can go through awaken one , have no possible path through other one ;

you wrote that you read powering up procedure .