DIY Audio Projects Forum

Hi all,

I'm working on a new project which happens to be a headphone amp. They are quick and easy and don't take up an ungodly amount of room. I looked at a couple of amp designs this weekend. Since I've already built a headphone amp that is pretty much a Cmoy amp with a class AB buffer output, I decided to try something different. The Szerekes amp looked fine but was just an output buffer. It has no gain and needs an extra stage to accommodate this. Luckily, MOSFETs can be direct-coupled with little problem.

I added an MPF102 to the input for a little gain. I didn't see much difference when switching to a J201 FET. The output devices is an IRF630 (I just heard several amp builders groan). I had them, therefore I used them. I intend on getting some better devices soon and transplanting them. If anyone can recommend a device I will take it into consideration. I have about 10 IRF820s in a box upstairs but don't know enough about them to use them.

Pictures abound!





The schematic is very simple and the component values were found by trial-and-error. I sort of modeled it after a 12AU7/MOSFET headphone driver. Since I don't have a laser printer, I use latex paint to map out the traces. Instead of Ferric Chloride, I found a formula for developing boards using Muriatic acid and hydrogen peroxide. There is a tutorial online somewhere detailing how to do this.





If the board I used were a bit bigger, I would have used some more power supply filtering and other useful add-ons. Perhaps I will be able to place them in the chassis elsewhere. Mainly the components I need are click-avoidance resistors.




The enclosure is a Fossil tin I scavenged from a thrift store for less than a dollar. It will do well to house the small amplifier. I am dreading the cutting of the rectangular hole for the power switch, but I have a nibbler which should take care of it.

Questions, concerns, comments, complaints, rants?
Ed

_________________

You can't have too much heat sink.

Nelson Pass would likely approve of this

Actually, the IRF6*0 isn't bad sounding, but other than the 610, the gates get a little hard to drive.

Cheers!

_________________
-= Gregg =-
* Ratings are for transistors - tubes have guidelines*
Home: GeeK ZonE
Work: Classic Valve Design

Hi all,

I finished the amp today and have it hooked up to my computer. It's to replace the Scott stereo receiver on the desktop (it occupies much less space). The amp is in its initial burn-in period and the material of choice is old AVGN episodes. None of the capacitors are special-the input caps are 220 nF mylars like the type sold at RS and the output caps are 470 µF electrolytic. I didn't bypass them with anything and everything still sounds fine. I don't believe in buying $18 "super caps" unless it's rated at tens of Farads.

I am not certain what I want to make next but I will post pics and data on this amp once it's been running a couple of days. Not sure about the effects of the so-called burn in but I remain open-minded for now. This brings me to another question...would it be beneficiary to burn-in MOSFETs or power transistors before using them in the system? I know that tubes are supposed to benefit from it due to their metal structure being affected by heat and stress, although I haven't observed the effect myself.

I'd have no qualms about burning in power-transistors as long as I could control the current and voltage to avoid overload. Burning in MOSFETs would make me nervous due to the possibility of damaging them with static electricity. Not that MOSFETs are particularly expensive but I haven't used anything but the cheap RS variety chips (IRF630 and IRF510). There are supposed to be 'audiophile quality' MOSFETs out there for about $10 or more apiece. They might have merit but for now I am not messing with them.

When this amp has run for a couple of days I plan to measure the voltages on the left and right channels and post them. I did notice that one channel runs slightly hotter than the other (one runs about 30 mA and the other runs closer to 50) but I can't tell a difference in audio quality. If anyone has questions, comments, answers, or the usual then please ask. If anyone wants to build this I'll try to be as helpful as possible.

I have never played with MOSFETs before but I like what I have seen. Perhaps I will use a couple to make a small speaker driver good for a few watts. It only takes 12.7 VPP into four ohms to get five watts.

Ed

_________________

You can't have too much heat sink.

I've let the amp play for a couple of days and have measured voltages. The voltages on the output look a bit low but the amp sounds great. I think it goes to show that one doesn't need an outrageous amount of idle current to get great sound. The amp sounds better than great-for some reason it has a calming effect after a tedious day at work.



The next time I build this type of amp, there will be a CCS to control the output current as well as provide a little bit of stability and efficiency. I'll also provide a pot in the drain circuit of the JFET (or tube?) and another CCS. The CCS will provide high AC impedance for more gain and the pot will allow an adjustable output voltage at the MOSFET's source. There will also be a resistor on the other side of the output cap that will allow discharging.

With the amp as-is, there is no hum nor are there pops. With the amp turned up fully there is some hiss but I believe changing the carbon film resistors with metal films might change that. I may not even bother with it as the amp sounds so good already.

I haven't run sine/square/triangle wave tests and I haven't tried to find a distortion figure. To me these don't matter as I like what I hear. I would like to get into servo and stability circuits so I could make more and more direct-coupled amplifiers. Such things are necessary to keep the bias points from wandering. I like the Gilmore amp but it seems a bit too complicated for me to work with right now.

Ed

_________________

You can't have too much heat sink.