Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Best value DIY headamp kit from Jaycar!!! (Pics)
New Posts  All Forums:Forum Nav:

Best value DIY headamp kit from Jaycar!!! (Pics)

post #1 of 17
Thread Starter 
Guys,

I have come across this post:

Rock Grotto Forum - Jaycar Headphone Amp Kit

It seems to be a good DIY kit for my HD650. I pickup the kits my local Jaycar store (the amp and the power supply) for about $50AUD...



2 hours later it is singing along... Wonderful sounding as it is. I followed the simple mod list from AK the improvement is very obvious the high is now well extended and the bass tighten up a little. There is plenty of driving power from this amp, volume can never go over 12 o'clock, in most case 9'clock will give my HD650 plenty of kick (Try that with another amp...)



As you can see from the pic I have further change the 10uf NP to 4.7uf metallised cap and upgraded the few 100uf electrolytic to better Elna (from RS).

At this point I have done a couple of blind folded AB comparison with a Solo borrowed from a friend. The result is at lease even handed (that is according to my friend who paid full retail for his Solo...)

Now, you know how happy I am. So far, included all upgrades this project is under $100 AUD.

At this point I have a couple of questions:

1. I have keep this amp on for hours during burn in, but the heat sink is very cool. Can someone please let me know if this amp is working in Class A or Class AB?

2. If it is working in Class AB how can I change to Class A? Any advantage?
post #2 of 17
you can determine the idle current by measuring for a voltage drop across one of the 4.7ohm resistors between the output transistors. Using ohms law you can figure out the current from the measured voltage.

V=I*R be sure to use proper units!
I=V/R

edited/added:
i cant see why the would specify *NOT* to use 1W resistors in the output positions. maybe they are afraid of inductance?
post #3 of 17
Your question:

1) Try measuring the bias. Check the voltage drop across the output resistors.

2) You can try increasing the diode current if you want to push it to Class A. Class A mainly has advantages like zero cross over distortion etc.



Oh and one thing to yell about the design, if the designer is going to use a relatively good CCS, why not use a diamond buffered design instead? It seems to me that it's quite a waste lol. You only have to replace the two diodes with two complementary driver transistors.
post #4 of 17
Quote:
Originally Posted by nikongod View Post
i cant see why the would specify *NOT* to use 1W resistors in the output positions. maybe they are afraid of inductance?
I think it was physically too large for the pad spacing and had to be tombstone mounted to fit. A smaller 1W resistor may do. Have to measure the spacing to see what size it is.
post #5 of 17
Replace input circuitry
  • Add volume control
  • Replace NP cap with 1K resistor
  • Replace 47k resistor with 470K resistor
  • Jumper 100R resistor
  • Skip 47pf Cap
post #6 of 17
Quote:
Originally Posted by FallenAngel View Post
Replace input circuitry
  • Add volume control
  • Replace NP cap with 1K resistor
  • Replace 47k resistor with 470K resistor
  • Jumper 100R resistor
  • Skip 47pf Cap
Replacing the cap with a resistor might not be such a great idea if the source has any DC offset.

Replacing that 47k resistor with a 470k resistor and jumpering the 100r resistor will likely result in a dramatic increase in DC offset with the LM4562.

And that 47pf cap aint hurting anything.

Edit: And wasn't this a magazine article a couple years ago?
post #7 of 17
Quote:
Originally Posted by ericj View Post
Replacing the cap with a resistor might not be such a great idea if the source has any DC offset.

Replacing that 47k resistor with a 470k resistor and jumpering the 100r resistor will likely result in a dramatic increase in DC offset with the LM4562.

And that 47pf cap aint hurting anything.

Edit: And wasn't this a magazine article a couple years ago?
Definitely not a good idea if source has DC, will sound better with source without DC offset.

Resistor on the input would work with the 470K as voltage divider and is just there to prevent ears blowing off if the wiper gets lifted off the (50K) pot.

100R+47pF looks like a high-pass filter.

Just wanted to make this look like a Pimeta/PPA style input
post #8 of 17
Quote:
Originally Posted by ericj View Post
Edit: And wasn't this a magazine article a couple years ago?
Yup... I made a post a year or two ago about it.

I still have not got round to listening to mine however, I got most of the way through replacing the electrolytics... replaced all with Panasonic FM and NHG, have some 6.8uF film I'll be using to replace the NP electrolytics on the input, but need to drill out new holes for them.

One issue I had with mine was the pcb mount RCA connectors wouldn't go into the board at all, I think the mounting holes on my board aren't right for them. No big deal though as I planned on soldering wires straight to the board anyway.

To the OP, are the changes worth it in your opinion ?
post #9 of 17
Quote:
Originally Posted by FallenAngel View Post
100R+47pF looks like a high-pass filter.
its a LOW-PASS filter. The cap sends HF into ground, LF gets by. this is something I would not remove without better reasons than were thought up for putting it in

The other stuff with adding a volume control, removing the input coupling cap, and replacing the 47kohm resistor with 470k looks cool as long as you stay away from opamps that are sensitive to input offset currents.

Im pretty sure this was designed to plug into a preamp or studio console that has the volume control functionality, but not headphone out. taking this thought 1 step further, the input low-pass network could be there to stop mirophopne/headphone feedback...

Quote:
Originally Posted by FallenAngel View Post
Definitely not a good idea if source has DC, will sound better with source without DC offset.
on teh assumption that this was designed as a bit of studio gear, studio people HATE DC coupling anything. seriously, if they could put a transformer or cap in between everything they would (and come pretty close to it). protecting the gear is more important than not having the sound of a film cap come through the average stereo anything that is recorded will be played on...
post #10 of 17
Thread Starter 
Quote:
Originally Posted by nikongod View Post
you can determine the idle current by measuring for a voltage drop across one of the 4.7ohm resistors between the output transistors. Using ohms law you can figure out the current from the measured voltage.

V=I*R be sure to use proper units!
I=V/R

edited/added:
i cant see why the would specify *NOT* to use 1W resistors in the output positions. maybe they are afraid of inductance?
Thank you for the tips Nikongod,

I have measured the voltage drop for all four 4.7ohm resistor and they read between 100mV to 50mV. Sorry for my ignorant, is that mean

0.1V / 4.7ohm = 0.021 A or 21mA? but that still not tell me if tha amp is working in Class A or Class AB.

PS: the stocked 1W resistor is 5% carbon. I have replaced with 1% metal film
post #11 of 17
Quote:
Originally Posted by FallenAngel View Post
Resistor on the input would work with the 470K as voltage divider and is just there to prevent ears blowing off if the wiper gets lifted off the (50K) pot.
The 47k reisistor to ground is there for more than that.

The lm4562 is a bipolar input opamp, and if you put it's input half a megohm from ground, the dc offset will increase by a wide margin.

Using the 'typical' properties of the lm4562 in tangent's dc offset calculator with a 47k resistor from noninverting input to ground results in a 1.65mv predicted dc offset. With 470k in that position it results in a predicted offset of 14.77mv, which is approaching uncomfortable territory.

If we use the 'maximum' values from the datasheet with a 470k resistor from noninverting input to ground, the DC offset could be as high a 99mv, and i I'm sure that most of us would greatly prefer not to push 99mv of DC voltage through a $300 headphone.

99mv is the worst-case scenario with the 470k resistor, but at 47k the worst-case is reduced to 11.8mv.

The lm4562 is an exceptionally good audio opamp, but it is *not a happy-go-lucky jfet part, and it must be used with care.
post #12 of 17
Thread Starter 
Quote:
Originally Posted by ericj View Post
Replacing the cap with a resistor might not be such a great idea if the source has any DC offset.

Replacing that 47k resistor with a 470k resistor and jumpering the 100r resistor will likely result in a dramatic increase in DC offset with the LM4562.

And that 47pf cap aint hurting anything.

Edit: And wasn't this a magazine article a couple years ago?
Yes Eric, this amp is nothing new 

Kit number KC5417 from Silicon Chip. Power supply kit KC5418. I have never thought the Jaycar kit is any good for serious audio application until now 
post #13 of 17
Thread Starter 
Quote:
Originally Posted by FallenAngel View Post
Replace input circuitry
  • Add volume control
  • Replace NP cap with 1K resistor
  • Replace 47k resistor with 470K resistor
  • Jumper 100R resistor
  • Skip 47pf Cap
[*]Add volume control (Done, a 50K pot is added to the input) [*]Replace NP cap with 1K resistor... Not sure about this.... will DC be a problem ?

What is the purpose of these mods? [*]Replace 47k resistor with 470K resistor[*]Jumper 100R resistor[*]Skip 47pf Cap
post #14 of 17
Thread Starter 
Quote:
Originally Posted by splaz View Post
Yup... I made a post a year or two ago about it.

I still have not got round to listening to mine however, I got most of the way through replacing the electrolytics... replaced all with Panasonic FM and NHG, have some 6.8uF film I'll be using to replace the NP electrolytics on the input, but need to drill out new holes for them.

One issue I had with mine was the pcb mount RCA connectors wouldn't go into the board at all, I think the mounting holes on my board aren't right for them. No big deal though as I planned on soldering wires straight to the board anyway.

To the OP, are the changes worth it in your opinion ?
Yeah Splaz, the RCA connector did not fit right. But no biggy there. I will get into some opamp rolling later after I finish with the rest of the mod
post #15 of 17
Thread Starter 
Quote:
Originally Posted by ericj View Post
The 47k reisistor to ground is there for more than that.

The lm4562 is a bipolar input opamp, and if you put it's input half a megohm from ground, the dc offset will increase by a wide margin.

Using the 'typical' properties of the lm4562 in tangent's dc offset calculator with a 47k resistor from noninverting input to ground results in a 1.65mv predicted dc offset. With 470k in that position it results in a predicted offset of 14.77mv, which is approaching uncomfortable territory.

If we use the 'maximum' values from the datasheet with a 470k resistor from noninverting input to ground, the DC offset could be as high a 99mv, and i I'm sure that most of us would greatly prefer not to push 99mv of DC voltage through a $300 headphone.

99mv is the worst-case scenario with the 470k resistor, but at 47k the worst-case is reduced to 11.8mv.

The lm4562 is an exceptionally good audio opamp, but it is *not a happy-go-lucky jfet part, and it must be used with care.
I am still using the stocked OPA2134. But from what you said I will try to get my hands on some LM4562... thanks mate.
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Best value DIY headamp kit from Jaycar!!! (Pics)