F5 Manual Amplifier Samples

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all!

I'd like to play with the sample circuits shown on F5 Manual on this weekend and so I'd like to make a few questions:

1) I saw one of them is designed for headphones and the other is a regular power amplifier, but I'd like to know why this is so and what would happen if I use the headphone amp to power a regular speaker and vice-versa.

2) Is there any risk of thermal runaway on these circuits? If yes, should I limit their operation time to avoid it or is there any other trick during operation?

The circuits are attached.

Thanks in advance
Daniel
 

Attachments

  • amp1.png
    amp1.png
    39.9 KB · Views: 562
  • amp.png
    amp.png
    30.2 KB · Views: 553
Last edited:
Heck,

go play with them, they are easy enough to switch around the parts or build both.
Be SURE to heatsink the Mosfets, they will get HOT.

Also, you do not need *exactly* the voltage shown... or exactly the cap values shown either.
Anything close will function, like ripped out of an old computer monitor or TV chassis? :D

If you test them both, the difference between them will be instantly apparent! :D
Do you have a scope and DVM?
Should have both if ur going to experiment with circuits.

Used scopes are pretty cheap these days.

_-_-bear
 
Hi all!

Firstly, thanks for everyone's answer! :)

AndrewT, I wrote F5 on the thread title because I found these samples at the beginning of the F5 manual, together with some basic explanations and examples about transistors/electronics etc.

I've simulated both on Multisim but just common drain seem to work. I'll post the common source output waveform here tomorrow to see if someone can help figuring out what goes wrong.
Also, what's the purpose of the 1k resistor between signal input and the input cap? Leaving it out removed some output attenuation in respect to input signal on simulation, but I don't know if it secure to do so in real life.

I'll try to build common drain tomorrow (let's see if my heatsink is enough) and give feedback.

Best regards
Daniel
 
Hey guys :)

I'm glad to say it worked.
I've attached some pictures of the build.

First try was with 2SK1530 on one channel and IRFP960 on the other. I've supplied it with 12V and got ~500mA bias on 2SK1530 and ~300mA on IRFP960.

I was expecting better audio quality from the Toshiba, but I couldn't perceive any difference between the two, at least with my headphone (Technics RP-DH1200). I got a lot of hum, I don't know if it was because trafo rating was too small (1A) or 4x4700uF caps weren't enough. I have also 4x0,47uF on the other side of the board.

Then I replaced 2SK1530 with another IRFP960 so that both channels share current equally and used a 90W (19,5V x 4,65A) Dell Laptop PSU to feed it still with the caps. Noise was completely gone (except if input is floating), and both output devices run at ~750mA bias.

I could get 1.5Wrms per channel, being feed from a M-Audio sound card with high level output, on 4R speakers with 10" Woofers. That gave me a nice listening level without distortion.

I've some questions:
1)Can I know whether the noise with the first PSU was caused by the trafo being too small or the caps being not enough?
2)Were I supposed to perceive any difference between the channel with Toshiba output and the other with IR output?
3)Can the caps on the audio path place any perceivable sonic degradation or not really?
4)Why do I get noise if input is disconnected and left floating instead of grounded?

After all thanks Papa for my first homemade amp. :cool:

Bye
 

Attachments

  • 2012-03-16_00-17-22_280.jpg
    2012-03-16_00-17-22_280.jpg
    600.6 KB · Views: 276
  • 2012-03-16_00-17-31_316.jpg
    2012-03-16_00-17-31_316.jpg
    519.2 KB · Views: 258
  • 2012-03-16_00-18-24_881.jpg
    2012-03-16_00-18-24_881.jpg
    514.3 KB · Views: 243
Last edited:
4.)
the terminated input wiring shows a low impedance to the input stage of the receiver.
The unterminated wiring shows a high impedance to the input stage of the receiver.

A low input impedance is much more resistant to picking up interference.
Or put another way, a high source impedance acts like an aerial.
 
1) No problem, measure and or listen to it all first, then change/add caps, or the trafo, and measure/listen to the differences. Yep, that could be expensive???
2) Yes. Objectively speaking, that is quite likely.
3) Ditto.
4) Yes, AndrewT is right. If you have any kind of radiated noise currents near your input, the high impeadance of an unterminated input will make lots of noise voltage at the input. Probably that noise responds to the gain in the circuit and now its 5-10 times or more worse.
Normally you would have only a few hundred ohms or less(due to the source impeadance of the previous stage). Any radiated noise signal near a low impeadance will not have enough power to develope a voltage on a low impeadance input. A typical high impeadance input is 100-1000 times higher load resistance/impeadance. Much easier to develope an artificial input voltage. Like an anttena! (areial)! Since you don't listen to an open input connector there are no worries! Most likely?
I think a possible flaw(not trying to be insulting or anything) in your thinking is that you seem to be using full monty amplifier transistor on a headphone amp??? It will sound a little better if you get the Iq up above an amp or so. But then, why don't you plug it into a speaker? You might consider a smaller transistor??? Or EUVL's DAO??? Maybe a variation of the BA-3 pre would suite your phones??? Just some suggestions...
:Pawprint:
 
Thanks for the explanations!

flg, actually I tried it with speakers. I've said it on the previous post. This is only a simple design I choose to start with and play a little. So nothing pretencious here. I want to try to bypass the output caps to see if I perceive any difference. What's a good value for them? 1/1000 or 1/10000 of the bigger ones?
 
Larger bypass is going to start working at a lower frequency.

So, something like a 10 to 20 ufd polyethylene (mylar) or better still a polypropylene, then a 0.1ufd of the same, and to be complete in all regards, then a 0.01ufd... likely you will hear little difference unless ur hearing is very good, the source is very clean, and the amp is properly biased into class A, and ur headphones are very good too. :D

Worth the experiment if you have caps hanging about - use whatever you have on hand to start. I salvage caps from old CRT computer monitor boards - good source for free parts of decent quality. While they still are around (fading away fast, they are).

_-_-bear
 
Hi bear

I tried to bypass the output caps with 1uF polystyrene, 1uF polyester and finally 100nF polyester (one at once) but I as you said I couldn't hear any difference at all. Tried one channel at once so that I could compare them... I tried also to substitute the input bipolar by 3x3,3uF oldschool Siemens polyester and 10x1uF polypropylene but again I couldn't hear any difference at all.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.