F5 power amplifier

Jump to Latest
#8,341
The bulb is not intended to work with two biased channels, not even one.. It is just a starter to see if evething OK.
To set the bias, disconnect the bulb.
If you set the bias with the bulb connected, you get a voltage drop in the rails.
If you then take the bulb out, rails voltage increases and so does the bias current.
 
Last edited:
#8,345
bobodioulasso said:
The bulb is not intended to work with two biased channels, not even one.. It is just a starter to see if evething OK.
To set the bias, disconnect the bulb.
If you set the bias with the bulb connected, you get a voltage drop in the rails.
If you then take the bulb out, rails voltage increases and so does the bias current.
Click to expand...

Hi bobodioulasso

I have never biased the amp with the bulb connected, just when checking each completed section. I only connected the bulb when both channels had been biased, and I was carrying out the final checks. It was at this point the bulb through me.

Anyway...better to be safe than sorry!

I always try to find the solution to a problem by reading-up before bothering you guys, especially AndrewT 😉

I really want to have a listen tonight but we are unfortunately going to a diner party, oh well tomorrow it is 🙄
 
#8,348
I have a pair of Peter Daniels F5 boards and a set of IR Mosfets, all ready to begin making my first discrete amp, and I am so totally daunted at the prospect of this 500 page tome...can anybody suggest a nice approximation of the advancements to the design (like the removal of part of the circuit some have raved about), or any commonly agreed parts choices, or a "show and describe your F5 here" page to get me on track?

Many thanks
Lucas
 
#8,352
Hi Folks

I have another question for you 😀

After biasing the amp yesterday I left it on for several hours with the lid off, and monitored the DC offset during this time. Before replacing the lid the offset rose to 10mV, but would hover around this value, occasionally slowly dropping to 6mv. After I replaced the lid the offset rose to ~23 mV dropping to around 17mV.

I know these values are low and are nothing to worry about but would you recommend re-trimming the bias again and replacing the lid? There are 6 ventilation holes top and bottom...12 in total, each hole has a diameter of 10mm. The heatsinks are hot, I tried the 5 second test but became bored after 30 :tilt:
 
#8,354
placing the trim pots (& PCBs) where they can be adjusted through the closed chassis would be far better.

Changes in mains voltage and changes in ambient temperature change the bias currents.
It's those changes in bias currents that show as output offset variations.

This is where the thermistors help. They give a bit of negative feedback to help attenuate the changes.
 
#8,355
AndrewT said:
placing the trim pots (& PCBs) where they can be adjusted through the closed chassis would be far better.

Changes in mains voltage and changes in ambient temperature change the bias currents.
It's those changes in bias currents that show as output offset variations.

This is where the thermistors help. They give a bit of negative feedback to help attenuate the changes.
Click to expand...

Thanks Andrew. I'll re-trim and re-close until I'm happy with the results. To be honest its probably worth doing when the amp is back in the rack as the study has been very hot today, especially with the amp running :hot:

I haven't omitted the mosfet thermistors on my amp.
 
#8,359
Electrolastic said:
Jackinnj:
Is that your new F5 layout? What are those 2 smaller power semi-conductors next to the Fairchild MOSFETs? No NTC thermistors? It looks interesting.
Click to expand...

It's my old F5 layout -- and the one I continue to use -- the smaller semi-looking devices are Caddock resistors. I subsequently replaced the Ohmite TW5's with Caddocks as well. The thermistors can't be seen from that view. I find that they lower the probability of headaches.
 
You must log in or register to reply here.
Top Bottom