Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Cavalli EHHA Embedded Hybrid Headphone Amp
New Posts  All Forums:Forum Nav:

Cavalli EHHA Embedded Hybrid Headphone Amp - Page 8

post #106 of 1813
5/4/09 at 9:49am
sachu I would like to reaffirm Ti's comments. I didn't realize what you were doing but even 2" lead length is a potential problem. I'd also recommend you put the fets onto the board. Their heatsinks are quite large and will handle any reasonable amount of power.
Reply
post #107 of 1813
5/4/09 at 10:43am
Quote:
Originally Posted by runeight View Post
Yes, he does, and has been having a hard time getting into the workshop.
Take your time, I'm a week away
Reply
post #108 of 1813
5/4/09 at 12:25pm
If you don't mind being the delay timer you can use one of these.
Reply
post #109 of 1813
5/4/09 at 2:23pm
Quote:
Originally Posted by amb View Post
Sachu, I just stumbled upon your build pics and have a comment about air-wiring the MOSFETs on the heatsinks. The inductance in the wires form an LC resonant network with the MOSFET gate capacitance, and could lead to instability (ringing square waves or even outright oscillation). The gate wiring to a MOSFET should be as short as possible and the gate stopper resistor should be immediately adjacent to the gate pin, this is why MOSFETs should be directly soldered on the PCB.

This is not related to your current problems but I thought you might want to be aware of this issue.
Quote:
Originally Posted by sachu View Post
Thanks for the info Ti. Will be keeping the length as short as possible (less than 2 inches).
Quote:
Originally Posted by runeight View Post
sachu I would like to reaffirm Ti's comments. I didn't realize what you were doing but even 2" lead length is a potential problem. I'd also recommend you put the fets onto the board. Their heatsinks are quite large and will handle any reasonable amount of power.
If it's too hard/too late to make the change and move away from air wiring, I suppose you could at least remove the gate stoppers from the board, and solder them directly to the mosfet pin, and then position the fets as close as possible. Just a thought...

But it is probably best to mount them on the board if possible.
Reply
post #110 of 1813
5/4/09 at 2:29pm
Its not possible to mount the heatsinks onto the board in the current setup.

The options I have are :

1> To remove the transformer and put it in a separate box, at least the S22 transformer.

2> Install a board underneath amp boards on as a chassis base and mount the S22 board there along with heatsinks

I am going to play around with placement a bit more and see what's possible..
Reply
post #111 of 1813
5/4/09 at 2:34pm
Sachu I'm sure you recall and TimJo's comment actually reminded me of when air wiring the MOSFETs on the starving student, to solder the resistor right up its leg/pin to prevent oscillation.
Reply
post #112 of 1813
5/4/09 at 3:06pm
I can at least atest to the EHHA being stable with air mounted mosfets - mine has been air mounted from the time I cased it up. My initial setup and test of the boards were with heatsinks on the board but once the amp was working, I cased it up and it has been going strong for the past 2 years.

If the S22 needs more stringency, you may be able to bolt the mosfets onto the chassis you have Sachu. We can stick the entire thing on the scope to see if there is any oscillation.

As for why your mosfets on the EHHA are not conducting, not sure - could you measure the gate voltage. The pin 6 opamp is ok - that is the offset correction, the 1.8V is higher than I recall..dB
Reply
post #113 of 1813
5/4/09 at 4:17pm
just wanted to break the nelson count so we can all stand on two feet now

Don, I put standoffs on the S22 board, need to drill 4 holes on the chassis so I can mount the MOSFETs directly onto the board.

I will report back with the gate voltages on the EHHA board in a few minutes.
Reply
post #114 of 1813
5/4/09 at 4:50pm
The voltages on the two MOSFETs ont he EHHA are

On the IRFZ34, Q9M,
the gate voltage is -1.36 ..but it varies from -1.36 to -1.45 volts.
The Drain voltage is at negative rail : -28.8 volts
Source is at 0 volts..less than 1 mV

On Q8M the IRFZ24, the gate voltage is +1.75 volts.
The Drain is at positive rail : +28.8 volts
Source is at 0 volts
Reply
post #115 of 1813
5/4/09 at 4:52pm
So what is the voltage at the output with respect to ground? And what are the voltages across the two 2R2 resistors?
Reply
post #116 of 1813
5/4/09 at 4:57pm
Alex, both are reading 0 volts..Output is at 0.4mV with respect to SG
Reply
post #117 of 1813
5/4/09 at 5:00pm
Then set the idle current to 100mA and see if it makes music.
Reply
post #118 of 1813
5/4/09 at 5:02pm
Alex, i thought you need more than at least 2V on the gate for the MOSFETs to conduct
Reply
post #119 of 1813
5/4/09 at 5:20pm
Took some measurements:

Q7 has
-1.33 on the emitter
+1.67 on the collector
-0.61 on the base.

Something clearly is wrong.

Q10,Q11 have +27.7 volts on the base
Q12, Q13 have -27.6 volts on the base
Reply
post #120 of 1813
5/4/09 at 5:20pm
I don't think the mosfets are conducting - this is similar to the scenario we had in the proto phase but we changed the trimmer to 1K and adjusted the bias resistor values and it came up fine - I was measuring around 3.2V at the gates ..dB
Reply
New Posts  All Forums:Forum Nav:
  Return Home

Head-Fi's Best Sellers

Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Cavalli EHHA Embedded Hybrid Headphone Amp