A lot of people are crazy about headphones. (check out HeadWize.com and Head-Fi.org.) Yet, the quality of the amplifiers used to drive headphones in portable devices and even some mainstream components can be very low. Sometimes this is due to cost savings, other times it is due to the need for portable devices to use a very low amout of power. So, if you want quality sound from your headphones, often it is necessary to get or build a headphone amplifier.
The "CMoy" headphone amplifier is one of the most popular designs. Its name comes from the person who published it- Chu Moy. The amplifier uses a dual opamp to drive headphones, it is very simple, and it can be made very small. Here are some links to get further information on this project:
I don't consider myself a "headphone fanatic" but look at this- I've actually got quite a few pairs of heaphones. I use headphones to prevent hearing loss while traveling on noisy airplanes. I grew up in the '80's and need all the help I can get! How good is the CMoy with these- lets see.
What the heck, I thought I'd build one. But only using parts I had around already. I did have to get the input/output connectors but had everything else laying around, including the case which I had gotten at the surplus store long ago. So my CMoy is built just like the basic one with some modification to the cirucit values as well as having a lot of flexibility and features.
Here is the schematic. This only includes one channel. The other is the same and powered by the same power supply.
Notes: You see here that I used some values different than the traditional CMoy. The goal was to make this out of parts that I already had on hand and not worry about the small things. For example, I did not have a 220uF capacitor to put between the 9 volt battery outputs, but I did have 2 small capacitors that were 10uF. Not as good some would say, but with a battery you don't really need capacitors here anyway. The resistor values may seem a little odd, but they are the measured values, not the values you would specify when purchasing. I did this by hand on a perfboard- old school style. I had to buy the TLE2426 chip, the RCA pcb connectors, and the headphone connectors, and the rest of the parts I scrounged up from things that I already had.
Notes: You see that i used an existing case and its PCB, which I removed all the parts from except the input DC connector and green LED. I mounted the connectors to the rear of the PCB upside-down and enlarged the rear panel cutouts somewhat. Once again, I'm tight on space. I only had red wire so used it for essentially all the connections. I hope this doesn't break, cause all the wires look the same. I mounted the input/output rear panel connectors upside down on a perfboard because I wanted to use the existing cut-out in the case, and this was the only way to have them fit. More importantly, I wanted to be able to take off the back panel and slide the entire amp out for potential repair and modification. This is why i used PCB mounted connectors instead of panel mount connectors (on the back). This amp disassembles easily, just unscrew the back panel, remove panel, and slide cover off.
Case and Connectors
Notes: Once again, I'm short on extra space in my "Trunk Interface II," now a CMoy headphone amplifier. Cannot believe I fit all that stuff in there. The power switch in the "up" position will run the CMoy on battery power. In the middle position the amplifier is turned off. In the "down" position, the amplifier is run on the wall-wart connection (or off if the wall-wart is not connected. Note to self- the middle is + and outer is -). There are no fuses in here- I didn't have room; I may try to add one later.
Notes: The first oscilloscope screen shows a punishing 10 ohm load at 20 Hz. Set to 0.2 v/div you see that here the CMOY is putting out more than 3 div, a little over 0.6 volts, which equates to 60 mA. At an easier frequency, 1 KHz, the CMOY is putting out .8 volts into 10 ohms, which is 80 mA. This easily puts out over 10V into high impedances like 400 ohms.
I thought that the virtual ground would limit me to about 32 mA of current but I'm getting over 60, which is pretty good, though I'm not sure why that is!