Hi, I'm building Helios speaker by Jeff Bagby, the minimum impedance Is about 4ohm, should I increase bias to 1,4A to reduce distortion for this load? Has anyone measured distortion on 4ohm load with higher bias?
Thank's
Luca
Thank's
Luca
Nelson's article on the F5m only shows distortion curves into 4 Ohms for 1A bias. He does mention that higher bias can reduce distortion.
It would be a simple experiment to try adjusting your F5m to higher bias. Your ears would be the best judge of whether that was a good match for your speakers.
It would be a simple experiment to try adjusting your F5m to higher bias. Your ears would be the best judge of whether that was a good match for your speakers.
One on the output watching DC offset, one across the N source resistor, the third on the P source resistor.....
Instructions from 6l6....note "source resistors" in text. In schematic they are the .47 dealios hanging off the outputs....
I'm shocked and shamed at myself
- It turns out that the two output transistors face the opposite direction compared to the image I've had in my head all along!
Which means that I haven't grasped an iota of how the circuit actually works during all this time
Once I've collected my thoughts again – I'll be back with some serious questions for someone to answer!!! 😡
(😉)
Always, always build in similar cases ..they all look the same....coming from a guy that hid two 5 cu ft jbl boxes in a 1000 sq ft bungalow!
Russellc
As you read through these threads, you will see me showing my total ignorance over and over. Zenmod is laughing. Regular whipper of Russellc! Only one way to learn, making mistakes!I'm shocked and shamed at myself
- It turns out that the two output transistors face the opposite direction compared to the image I've had in my head all along!
Which means that I haven't grasped an iota of how the circuit actually works during all this time
Once I've collected my thoughts again – I'll be back with some serious questions for someone to answer!!! 😡
(😉)
Russellc
Both of the output mosfets operate in the common source mode. This allows the output stage to have voltage gain.
The mosfet sources (with source resistors) are at AC (signal) ground potential, in this case the positive and negative power supply rails.
And the two connected-together drains are the signal output, with signal return to ground.
Any reason why a bleeder circuit with a LED was not part of the power supply design for this amp?
Yes, it’s a belt and suspenders thing, but in a Class A, heavily biased amp as it is here (1A to 1.4A), those power supply capacitors will discharge quickly while IN circuit. If you’ve charged them up without connecting it to the main F5m circuit, they can stay charged and present a hazard to the diy’er. So be careful. And smart. So basically…it’s a choice you can add.
Note that the capacitors used are only 4700uF in size, so they are much smaller than typically used and as such, the energy in joules for each capacitor will be smaller as well. He uses 6 of them for each rail to get a decent capacitance value. 4 in parallel are used prior to the thermistor and 2 are used after. So 18,800uf -> Thermistor -> 9400 uf.
You can always add a bleeder or two, no big deal. And an LED or two, again no big deal. 20K-25K is a decent value to use (with 24V rails) in series with an LED, if you need one for a light show. And you can always use a calculator to figure out the perfect resistor value based on the characteristics of the LED (2-5mA is plenty for most LEDs) you want to use. Or place a pot and have more fun!
Or…you can just use the V3 diyaudio power supply at the diyaudio store. This includes bleeders and LED options. Use the schematic and BOM from the V3 supply as guides for any modifications of the F5M supply.
The F5M PS presented by NP on the F5m is simpler (in my view), with some interesting quirks like using a thermistor as the “R” in a CRC supply, etc….It’s actually a smart design because he not only includes the thermistor before the primary (which serves as the primary softstart) but also the thermistor after the 1st capacitor which is a ‘secondary‘ softstart for a nice slowish warmup - inrush current is suppressed in 2 parts of the supply - cool!
Thermistors have high resistances when cold at 25 degrees C. Once they warm up (which they will with a 1A-1.4A current draw), the resistance drops considerably. In example, the thermistor recommended is 10 ohms and drops to <1 ohm typically in operation, which is perfect as the “R” in a CRC supply (check out the graph in the link above). I bet when the amp is turned off, the supply voltage drop will take a little time since the thermistor has to cool - of course this can evaluated on the bench and bleeders added if need be.
Best,
Anand.
Last edited:
Anand, thanks. The reason I asked is that I did in fact test the PS before connecting the amps and now it’s in a charged state. I was going to connect a light bulb to the outputs of the PS. Would this be an ok method to discharge the caps?
With dual power supply setups like most first watt amps have, do both sides need drained to ground, or will one side drain both, right?
Thanks,
Russellc
Thanks,
Russellc
If you are going through the trouble of adding bleeder resistors, go ahead and add one bleeder for V+ and one bleeder resistor for V-. Belt and suspenders approach. See schematic of V3 diyaudio Universal Power Supply. You’ll notice R9 and R10 there for that reason.
Best,
Anand.
Best,
Anand.
I just drained it off with a lamp socket and an incandescent bulb. Worked pretty quickly. Once the amp boards a connected, I suppose it’s a none issue moving forward.
No bleeders, I meant to manually discharge with 100r resistor. My power supplies that have bleeders have them on both sides, so I figured that both sides of the supply should both be discharged to ground. Oh, and asking as to First Watt style power supplies, not necessarily this one. After blowing a chunk out of a pair of snippers on a powered up supply, I have become cautious! 😔
Russellc
Russellc