Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › mini^3 left channel really hot
New Posts  All Forums:Forum Nav:

mini^3 left channel really hot

post #1 of 17
6/30/10 at 9:28pm
Thread Starter 

I just finished building my AMB mini^3 today and it passed all the initial checks. When I plug in my headphones, the right channel works fine but the left channel produces no sounds and gets really hot. I also notice that the U5 gets really hot as well. so I did a recheck and here's what I got so far

 

Edit: the quiescent current check reads 97mA drops by 0.1mA every 2sec or so. I'm using the high performance amp

edit2:The rail splitter check produced 0mV on V- and V+ as well as 0mV on V- and IG

DC offset:

OG: 1.4mV

OL: 3.7V

OR: -4.2mV

 

Can anyone help me out?

Quiescent current check


Edited by toyboof - 7/1/10 at 12:03pm
Reply
post #2 of 17
6/30/10 at 9:54pm

What's the DC offset on all the channels?

Reply
post #3 of 17
6/30/10 at 11:06pm
Thread Starter 

OG: 1.4mV

OL: out of range

OR: -4.2mV

Reply
post #4 of 17
6/30/10 at 11:39pm
Quote:
Originally Posted by toyboof View Post

OG: 1.4mV

OL: out of range

OR: -4.2mV


Out of range? Select anonther range to produce a reading!

You could try to reflow the solder joints on U5. U5 could be blown. 

Reply
post #5 of 17
7/1/10 at 6:14am
Thread Starter 
Quote:
Originally Posted by Beftus View Post



Out of range? Select anonther range to produce a reading!

You could try to reflow the solder joints on U5. U5 could be blown. 


OL: 3.7V

Reply
post #6 of 17
7/1/10 at 8:23am

Please tell me you didn't plug in expensive headphones with a 3.7 volt DC offset. That's a great way to destroy any headphone.

 

The DC offset should obviously be as close to zero as possible, something sounds seriously wrong with the left channel. I'm actually quite surprised that there is no capacitor in the signal path output. I would have put one there for safety reasons, and if the builder knows what they are doing and the circuit seems to be working with a close to zero DC offset before the capacitor it could be safely shorted out which would probably result in lower distortion.

 

I would double check all of your resistor color codes and make sure the correct values are installed - especially important are R1L/R, R3L/R, and R4L/R. Don't try to measure it in-circuit with a multimeter, the reading will be inaccurate.

Reply
post #7 of 17
7/1/10 at 10:45am

U5 is likely blown. Quiescent current draw is way to high (hence the high temperature of U5). A properly working hi-perf Mini3 draws approx 25 mA. Mine draws 24.2mA.


Edited by Beftus - 7/1/10 at 10:55am
Reply
post #8 of 17
7/1/10 at 12:00pm
Thread Starter 
Quote:
Originally Posted by natrix View Post

Please tell me you didn't plug in expensive headphones with a 3.7 volt DC offset. That's a great way to destroy any headphone.

 

The DC offset should obviously be as close to zero as possible, something sounds seriously wrong with the left channel. I'm actually quite surprised that there is no capacitor in the signal path output. I would have put one there for safety reasons, and if the builder knows what they are doing and the circuit seems to be working with a close to zero DC offset before the capacitor it could be safely shorted out which would probably result in lower distortion.

 

I would double check all of your resistor color codes and make sure the correct values are installed - especially important are R1L/R, R3L/R, and R4L/R. Don't try to measure it in-circuit with a multimeter, the reading will be inaccurate.

 

When I first turned them on, I tried with some cheap dollar store kind, so luckily my good headphones were saved.

I doubled checked all my resistors before installing them, plus I got all my parts from Glass Jar audio as a kit, so i believe the resistors/capacitors are correct.


 

Quote:
Originally Posted by Beftus View Post

U5 is likely blown. Quiescent current draw is way to high (hence the high temperature of U5). A properly working hi-perf Mini3 draws approx 25 mA. Mine draws 24.2mA.

 

Alright, I'll order in a new U5. Is there anything wrong with both my rail splitter values being zero? I edited my first post a bit with some updated numbers. Also, is there anything wrong with a negative right DC offset?
 


Edited by toyboof - 7/1/10 at 12:06pm
Reply
post #9 of 17
7/1/10 at 12:22pm

The rail splitter is used for generating a virtual ground to provide a return path for the left and right channels, right...? If it was completely broken, it would have a DC level which was far from input signal ground, "IG". Another good way to test it is to measure the voltage between "OG" and "V+" along with "OG" to "V-". The numbers should be identical, but opposite.

 

A couple millivolts of DC offset is nothing to worry about. It should be exactly zero, but whatever. Also, I would be measuring the DC offset of the left and right channels with "OG" as a the "ground" for your multimeter as that is what your headphones will be using as a ground.

Reply
post #10 of 17
7/1/10 at 12:38pm


Are you sure about "the rail splitter check" because that doesn't add up.  
 

Quote:

 

Edit: the quiescent current check reads 97mA drops by 0.1mA every 2sec or so. I'm using the high performance amp

edit2:The rail splitter check produced 0mV on V- and V+ as well as 0mV on V- and IG

DC offset:

OG: 1.4mV

OL: 3.7V

OR: -4.2mV

 

Reply
post #11 of 17
7/1/10 at 1:56pm
Thread Starter 
Quote:
Originally Posted by natrix View Post

The rail splitter is used for generating a virtual ground to provide a return path for the left and right channels, right...? If it was completely broken, it would have a DC level which was far from input signal ground, "IG". Another good way to test it is to measure the voltage between "OG" and "V+" along with "OG" to "V-". The numbers should be identical, but opposite.

 

A couple millivolts of DC offset is nothing to worry about. It should be exactly zero, but whatever. Also, I would be measuring the DC offset of the left and right channels with "OG" as a the "ground" for your multimeter as that is what your headphones will be using as a ground.

measuring voltage between OG and V+:

With the mini^3 turned on, without the battery, black probe on OG and red probe on V+, I get -0.1mV.

With the same setup, but this time red probe on V-, I still get -0.1mV

 

with the amp turned on and battery in, I re-measured the output DC offset:

with my black probe on "OG", I get:

IR:-1.3mV

OL:3.65V

OR:-5.4mV



 

Quote:
Originally Posted by MisterX View Post


Are you sure about "the rail splitter check" because that doesn't add up.  
 

I just double checked "the rail splitter check" again and it's still the same values. Do you know if I can mail this amp to a professional to have it looked at? it's starting to drive me nuts.
 

Reply
post #12 of 17
7/4/10 at 10:45am


 

Quote:
Originally Posted by MisterX View Post


Are you sure about "the rail splitter check" because that doesn't add up.  
 





Quote:
Originally Posted by toyboof View Post

Edit: the quiescent current check reads 97mA drops by 0.1mA every 2sec or so. I'm using the high performance amp

edit2:The rail splitter check produced 0mV on V- and V+ as well as 0mV on V- and IG

DC offset:

OG: 1.4mV

OL: 3.7V

OR: -4.2mV

 

Can anyone help me out?

Quiescent current check

I would agree with MisterX. If you are getting 3.7V on OL to IG, V- to V+ cannot be 0mV. It should be your battery voltage (9V). V- to IG should be 1/2 of that. Also, your further tests with the battery out are meaningless.

 

Carefully go through the entire initial check and when you find a deviation, report back what you found.
 

Reply
post #13 of 17
7/4/10 at 11:57am

I'm also confused by your rail-splitter check.  This check must be done with the battery installed.  As a general rule, you need to have a power source if you are going to measure voltages in a circuit.  The rail-splitter is checked by measuring voltages  between V+, V- and IG.  IG should be at the halfway point between V+/V-.

 

Quote:

measuring voltage between OG and V+:

With the mini^3 turned on, without the battery, black probe on OG and red probe on V+, I get -0.1mV.

With the same setup, but this time red probe on V-, I still get -0.1mV

 

 

Your 3.7 V offset is exactly what happened to me when I had a fried U5 (twice).  I agree with others, that's most likely what's happened here.  U5 is a slight pain to replace- I cut the leads as gently as I could with an exacto knife and then cleaned things up, as mentioned in AMB's directions on the Mini^3 page.

 

Good Luck!

Reply
post #14 of 17
7/6/10 at 7:28am
Thread Starter 
Quote:
Originally Posted by hiker101 View Post

I'm also confused by your rail-splitter check.  This check must be done with the battery installed.  As a general rule, you need to have a power source if you are going to measure voltages in a circuit.  The rail-splitter is checked by measuring voltages  between V+, V- and IG.  IG should be at the halfway point between V+/V-.

 

 

Your 3.7 V offset is exactly what happened to me when I had a fried U5 (twice).  I agree with others, that's most likely what's happened here.  U5 is a slight pain to replace- I cut the leads as gently as I could with an exacto knife and then cleaned things up, as mentioned in AMB's directions on the Mini^3 page.

 

Good Luck!

Ahh, I must've mis-read AMB's instructions on the initial check. I'm new to the DIY audio so please pardon my newbie mistakes. I checked the rail splitter this time with battery in and received 7.3V and 3.6V for my V+ to V- and V- to ground, respectively. So I guess my rail splitter voltages are correct. I just placed an order for another U5 chip, It should be coming in shortly. I'll be sure to keep this updated.

 

Out of curiosity, how did you fry your U5 chip? was it similar to mine, where after its turned it on, everything becomes messed up?
 


Edited by toyboof - 7/6/10 at 7:30am
Reply
post #15 of 17
7/6/10 at 9:27am

.


Edited by ShinyFalcon - 7/6/10 at 9:29am
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › mini^3 left channel really hot