Greetings to everyone,
I wanted to make an amplifier with LM4702, which has been in my drawer for 15 years. Previously I made 3 separate LM4702 amplifiers and I liked them too much. However, I gave some of them to my friends and I don't know where some are.
I also had Alfet Lateral ALF08NP MOSFETs waiting for at least 10 years in the same drawer. So I guessed that the two would be a beautiful duo.
Since I know that a buffered chipamp will give better SNR-THD performance according to my previous experience. I designed the circuit as buffered.
PCB design was quite simple, as Alfet MOSFET is P and N channel is in the same case.
I wanted to try the circuit without the buffer in the first run.
I simply installed it and assembled it on a single heatsink. Bias trimpots were approximately at mid position, while approximately 300ma per channel was formed. Then I slowly adjusted to 125mA/channel. When the MOSFETs heat up to about 40 degrees (about 10 minutes), I saw that the biases remained constant. And I measured the output noise, inpus were shorted: I read 30uv (20mhz bandwidth) noise in my Oslioscope.
Then I went to the listening test. Connected my Arylic HD-DAC (opamps replaced with OPA1656) to my new amplifier and I have been listening to Wharfedale Diamond 9.6 speakers for hours.
I have to say clearly that although it is not buffered for now, the amplifier's performance surprised me again. I must say that it is a very clean noiseless and distortion -free amplifier. Since I have listened to plenty of Class D amplifier lately, the clarity of the device is quite impressed me too much.
Although both LM4702 and lateral MOSFETs are no longer produced, I would recommend it to anyone who has the opportunity.
I wanted to make an amplifier with LM4702, which has been in my drawer for 15 years. Previously I made 3 separate LM4702 amplifiers and I liked them too much. However, I gave some of them to my friends and I don't know where some are.
I also had Alfet Lateral ALF08NP MOSFETs waiting for at least 10 years in the same drawer. So I guessed that the two would be a beautiful duo.
Since I know that a buffered chipamp will give better SNR-THD performance according to my previous experience. I designed the circuit as buffered.
PCB design was quite simple, as Alfet MOSFET is P and N channel is in the same case.
I wanted to try the circuit without the buffer in the first run.
I simply installed it and assembled it on a single heatsink. Bias trimpots were approximately at mid position, while approximately 300ma per channel was formed. Then I slowly adjusted to 125mA/channel. When the MOSFETs heat up to about 40 degrees (about 10 minutes), I saw that the biases remained constant. And I measured the output noise, inpus were shorted: I read 30uv (20mhz bandwidth) noise in my Oslioscope.
Then I went to the listening test. Connected my Arylic HD-DAC (opamps replaced with OPA1656) to my new amplifier and I have been listening to Wharfedale Diamond 9.6 speakers for hours.
I have to say clearly that although it is not buffered for now, the amplifier's performance surprised me again. I must say that it is a very clean noiseless and distortion -free amplifier. Since I have listened to plenty of Class D amplifier lately, the clarity of the device is quite impressed me too much.
Although both LM4702 and lateral MOSFETs are no longer produced, I would recommend it to anyone who has the opportunity.
Attachments
Sercan85 will make a clone of this and will measure.. I haven't got proper instrument for FFT etc measurements.. I guess in next couple weeks he will share the measurements..