From: "Saved by Internet Explorer 11" Subject: DIY 12AU7 (ECC82) Tube - IRF612 MOSFET Hybrid Headphone Amplifier Date: Mon, 21 Apr 2014 16:01:00 -0700 MIME-Version: 1.0 Content-Type: multipart/related; type="text/html"; boundary="----=_NextPart_000_0000_01CF5D7A.E37C4840" X-MimeOLE: Produced By Microsoft MimeOLE V6.1.7601.17609 This is a multi-part message in MIME format. ------=_NextPart_000_0000_01CF5D7A.E37C4840 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Location: http://diyaudioprojects.com/Solid/Tube-Mosfet-Hybrid-Headphone-Amp/
Rogers Gomez |
=20 | =20 |
UPDATE -=20 12 April 2010 - This project has been superseded by the NP= -100v12:=20 diy 12AU7 (ECC82) tube / IRF510 MOSFET headphone amplifier project = page.=20
I have always been intrigued by tube amplifiers, but most DIY kits = are very=20 expensive and use very high voltage. So I decided to build an amplifier = that=20 would be inexpensive and had the least amount of parts necessary to = drive a pair=20 of 32 ohm Grado headphones.
Having built several YAHA amps based on the fa-schmidt design, and a Szekeres Mosfet follower I wondered how the two = would sound=20 together. So I built the schematic into TINA-TI, a free spice based = program to=20 test circuits before the build, and the results were remarkable. Nearly = 20dB of=20 gain across 20Hz-100kHz from a 13VDC power supply.
As you see in the schematic and parts list, there are less than 30 = discrete=20 components and most DIY'ers will have them as spares from other builds. = I chose=20 the 12AU7 / ECC82 vacuum tube because it can be driven with low voltage = and the=20 filament voltage is 12.6 volts, so there is no need to regulate the = voltage any=20 further. I used 1/4W resistors in the first stage and 2W in the second. = The 2W=20 resistors may be overkill but I did not want to change them later. The = 20ohm=20 resistor must be a minimum of 5W and do not use wire wound, as the = inductive=20 characteristics will distort the response curve.
QTY | Label | Value |
---|---|---|
2 2 2 2 2 2 2 2 = 2 2=20 2 1 2 1 1 1 |
P1 C1 C5 R1 C2 R5 R4 = R3=20 C4 R2 R6 VT1 T1 -- -- -- |
100k 100nF 100nF 1M 1uF = 20ohm 5W=20 200 ohm 2W 220k 2W 470uF 50V 47k 5k 2W 12AU7 / ECC82 IRF612 = LED Holder=20 LED 9-pin Tube Socket |
Additional Notes - Parts List:
For the enclosure I used a Lansing MicroPak "C"; which I found on = eBay for=20 approx $8. This item was an overstock of some surround decoders and came = with=20 RCA connectors and pre-punched faceplates. The tube socket should be a = 9 pin,=20 this one was for older pixie tubes so there are more holes, but the = 9-pin match=20 perfectly.
I used RCA connectors for input and a 1/4" jack for an output, this = was to=20 accommodate my Grado SR125's and the holes in the plates were punched = for a 1/4"=20 jack.
The PCB had pre-existing solder runs and a ground bus so this = dictated the=20 location of the components If you build this on proto board make sure to = leave=20 room for the heatsink and the 20 ohm resistors to dissipate heat. Make = sure to=20 insulate the Mosfet from the heatsink with mica, and also use heatsink = compound=20 for better heat transfer. The tab of the Mosfet is at 12-13 volts and = will=20 destroy the power supply and component if it shorts to ground.
After all of the components were soldered I smoke tested the unit and = made=20 any necessary corrections.
The tube heater draws 150mA at 12.6 and the whole amp draws 580 mA at = startup=20 and stabilizes at 550mA after the tube warms up; which takes a few = seconds. I=20 used a Canon AD-360U switching power supply from their small bubble jet = printer.=20 Since the drain of the Mosfet is connected directly to the positive rail = any=20 noise will be amplified. This power supply had virtually no noise.
I used a 12V muffin fan dropped to 9V to cool the amp, two holes in = the top=20 allow air to rush in and the existing holes in the front and rear plate = allow=20 the air to flow well.
Here are a few scope shots of the 10 Hz sine wave response, as well = as the=20 square wave response at 100 Hz. The performance was very consistent and = the=20 voltage stayed the same over the most of the sweep. The TINA curve = shows approx=20 19dB of gain over the spectrum of the audible range. The sound from my = Grado=20 SR125 headphones is crisp with a tight lower end. This amp is very nice=20 considering the low voltage and low component count. It may not be true=20 audiophile quality, but the average cost is only $40 USD or less.
This amp is perfect for the novice builder and the components are = available=20 at mouser, digikey. You can substitute other Mosfets as long as the = resistance=20 is similar and the internal capacitance does not effect the response = curve. It=20 is best to use TINA-TI to build the schematic and make any = changes, this=20 way you can check the ac output before building. This will save you much = time. =20
Lastly, make sure that your headphones are not plugged into the jack = while=20 powering up or down, this amp as well as other DIY builds have a large = rush of=20 current at power up and down and damage can occur to you headphones if = you are=20 not careful.
I am sure you will be pleased with the sound of this little amp and = better=20 yet the price.
UPDATE - 28 August 2008 - I have built another version of this = amplifier which is smaller, has less parts and runs off of a 12V = battery. For=20 full details, see the 12AU7=20 Tube / IRF612 MOSFET Hybrid Headphone Amplifier update on the DIY = Audio=20 Projects Forum.
UPDATE -=20 12 April 2010 - This project has been superseded by the NP= -100v12:=20 diy 12AU7 (ECC82) tube / IRF510 MOSFET headphone amplifier project = page.=20
=20=20 |