Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Bijou All Tube Futterman Headphone Amplifier
New Posts  All Forums:Forum Nav:

Bijou All Tube Futterman Headphone Amplifier - Page 119

post #1771 of 3981
11/4/08 at 12:17am
That's good!! Mostly.

The B+ is set to 250V??

This would mean that the pass mosfet is dropping about 60V at 70mA which will burn a bit over 4W. How hot is the pass mosfet?

Also, what is the voltage at the R8 point?

Are you using EZ81s?

Thanks. You are very patient with my questions.

If the mosfet is getting too hot we can fix this with a few resistor value changes once I know what the various voltages are.
Reply
post #1772 of 3981
11/4/08 at 1:24am
Quote:
Originally Posted by runeight View Post
That's good!! Mostly.

The B+ is set to 250V??

This would mean that the pass mosfet is dropping about 60V at 70mA which will burn a bit over 4W. How hot is the pass mosfet?

Also, what is the voltage at the R8 point?

Are you using EZ81s?

Thanks. You are very patient with my questions.

If the mosfet is getting too hot we can fix this with a few resistor value changes once I know what the various voltages are.
no worries. i've taken a lot from the generous DIY community - if the Dude can give back, the Dude abides.

i powered down before going out to pick up a 12-pack for a pre-celebration (hopefully) on this election eve.

so, with 20 min of warm-up i pegged these numbers on the Fluke:

Q2 = 73 C
Q1 = 45 C
TIP50 = 26 C
R8 = 320V

B+ set to 250V without amp boards attached - only 100kR.

yes, EZ81 - and ECC99 + 6N1P on the amp.

are my DAZ 370s up to the task? i'm picturing Scotty telling Captain Kirk, "I'm giving her all she's got!".
Reply
post #1773 of 3981
11/4/08 at 1:36am
Your transformers should be well up to the task.

But, it's Q2 where the problem could be. So when you fire up with amp boards in place pay attention to the temp of Q2. Don't let the PS run for very long before measuring it.

I think it's going to get too hot. The two ways to adjust for this are to use EZ80s and to replace the 4k7s with 10k resistors. This will drop the voltage at the R8 point and at the same time burn more power in the BJTs.

But, first the test . . . after you celebrate.
Reply
post #1774 of 3981
11/4/08 at 1:45am
Quote:
Originally Posted by runeight View Post
Your transformers should be well up to the task.

But, it's Q2 where the problem could be. So when you fire up with amp boards in place pay attention to the temp of Q2. Don't let the PS run for very long before measuring it.

I think it's going to get too hot. The two ways to adjust for this are to use EZ80s and to replace the 4k7s with 10k resistors. This will drop the voltage at the R8 point and at the same time burn more power in the BJTs.

But, first the test . . . after you celebrate.
what test?

Q1 and Q2 are rated for 150 C, are they not? MOSFETs have a negative thermal coefficient? you obviously know more than i do - could you expand?

the PS boards have been running for a total of 1 hr.

i'm still pegging around 73 C on Q2 after 45 min.
Reply
post #1775 of 3981
11/4/08 at 3:14am
Pabbi, I was hoping they would get it delivered today. Well, OK we'll see what tomorrow brings.
Reply
post #1776 of 3981
11/4/08 at 3:19am
Quote:
Originally Posted by amphead View Post
Pabbi, I was hoping they would get it delivered today. Well, OK we'll see what tomorrow brings.
Reply
post #1777 of 3981
11/4/08 at 12:19pm
Quote:
Originally Posted by fishski13 View Post
what test?

Q1 and Q2 are rated for 150 C, are they not? MOSFETs have a negative thermal coefficient? you obviously know more than i do - could you expand?

the PS boards have been running for a total of 1 hr.

i'm still pegging around 73 C on Q2 after 45 min.
Yes, but I'm still confused. Is the 73C reading with just a 100k load or with the full amp?
Reply
post #1778 of 3981
11/4/08 at 1:17pm
Quote:
Originally Posted by amphead View Post
Pabbi, I was hoping they would get it delivered today. Well, OK we'll see what tomorrow brings.

Eagle has landed, but I chose to machine out the top last night (safety first - I have too many mobile creatures not to)... and, as always, additional challenges. To be fair, i only spent about 15 minutes with it last night, and will be more diligent this evening.

I am afraid, in the end, the balanced output of the Cambridge Azur 840c is simply too high for the levels of attenuation... but, more observation is required.
Reply
post #1779 of 3981
11/4/08 at 2:05pm
Quote:
Originally Posted by runeight View Post
Yes, but I'm still confused. Is the 73C reading with just a 100k load or with the full amp?
powering 2 amp boards.
Reply
post #1780 of 3981
11/4/08 at 4:00pm
OK. Now this makes sense.

Yes, the mosfets are 150c devices. Keeping them below 75C is generally good practice, but even 73C means that the junction temp is quite a bit hotter than that.

So what is the R7/R8 voltage when the amps are powered?

In the original design, I set the conditions so that R7 and R8 would dissipate a bit more than 1W. This then put the mosfet at about 1.5W

But the cap mulitpliers don't drop as much voltage as the passive resistors so this means that more voltage is being dropped across the mosfet which leads to higher power dissipation.

What's happening, based on your previous numbers, is that the mosfet is dropping about 60V at 70mA which is about 4W. This is really more than the heatsinks were designed for in the original design where R7 an R8 took up a lot of this power.

But, since the BJT cap mulitpliers have very little voltage drop across them we are seeing this across the mosfet.

I would suggest that it would be better to increase the 4k7s to about 10k. This will put about 1W on each of the BJTs and about 2W on the mosfet. It will be much happier at this power dissipation.

Of course, I'm using voltage measurements that you sent before and I'm not sure whether they were measured with or without the amps attached.

Does this make sense?
Reply
post #1781 of 3981
11/4/08 at 5:43pm
Quote:
Originally Posted by runeight View Post
OK. Now this makes sense.

Yes, the mosfets are 150c devices. Keeping them below 75C is generally good practice, but even 73C means that the junction temp is quite a bit hotter than that.

So what is the R7/R8 voltage when the amps are powered?

In the original design, I set the conditions so that R7 and R8 would dissipate a bit more than 1W. This then put the mosfet at about 1.5W

But the cap mulitpliers don't drop as much voltage as the passive resistors so this means that more voltage is being dropped across the mosfet which leads to higher power dissipation.

What's happening, based on your previous numbers, is that the mosfet is dropping about 60V at 70mA which is about 4W. This is really more than the heatsinks were designed for in the original design where R7 an R8 took up a lot of this power.

But, since the BJT cap mulitpliers have very little voltage drop across them we are seeing this across the mosfet.

I would suggest that it would be better to increase the 4k7s to about 10k. This will put about 1W on each of the BJTs and about 2W on the mosfet. It will be much happier at this power dissipation.

Of course, I'm using voltage measurements that you sent before and I'm not sure whether they were measured with or without the amps attached.

Does this make sense?
yes, perfect sense.

all measurements were done on the PS with the amp boards connected.

i should have new readings with 10K replacing 4k7 shortly.
Reply
post #1782 of 3981
11/4/08 at 5:46pm
OK. Please measure the R7/R8 points when your're finished.
Reply
post #1783 of 3981
11/4/08 at 6:25pm
Quote:
Originally Posted by runeight View Post
OK. Please measure the R7/R8 points when your're finished.
amp boards connected, with 10 min. or so, here's what i get:

R8 point = 332V
R7 point = 308V
Q2 = 68 C
Q1 = 40 C
TIP50 = 30 C

so if my maths are correct, the Q2 mosfet is dissipating around 1.7 W

my mosfets are not isolated with pads from heatsinks - only a thin layer of thermal grease.
Reply
post #1784 of 3981
11/4/08 at 8:14pm

'Blue' update

Here are some new pix of my 'Blue' Bijou showing all the updates/mod's that
I've made since it was built.

http://i183.photobucket.com/albums/x...d/100_1521.jpg
http://i183.photobucket.com/albums/x...d/100_1517.jpg
http://i183.photobucket.com/albums/x...d/100_1520.jpg
http://i183.photobucket.com/albums/x...d/100_1516.jpg
http://i183.photobucket.com/albums/x...d/100_1518.jpg
http://i183.photobucket.com/albums/x...d/100_1509.jpg
http://i183.photobucket.com/albums/x...d/100_1510.jpg
http://i183.photobucket.com/albums/x...d/100_1511.jpg
http://i183.photobucket.com/albums/x...d/100_1513.jpg
http://i183.photobucket.com/albums/x...d/100_1514.jpg
http://i183.photobucket.com/albums/x...d/100_1515.jpg
http://i183.photobucket.com/albums/x...d/100_1519.jpg

Here's what I've done so far:
Added a separate heater tranny for the PS.
Regal's mod's.
Amphead's mod's.
Choke PS mod, variation 2.
6H30 gold pin input tubes.
ECC99 gold pin output tubes.
EZ81 PS rectifier.
Auricaps in C3, C4 and C6.
Changed signal wiring to cryo'ed Vampire (buried under all that shielding).
Stepped attenuator (50K) from ebay (Taiwan).

I just installed the stepper yesterday, and now understand the enthusiasm for them. The sum total of all the mod's has produced one fantastic listening experience. It's very detailed and dynamic, yet not at all fatiguing.

Thanks, Alex, for all your hard work on this design, and your continued efforts
in this thread. Thanks, also, to all of you who continue to tweak this great
design. I finally think I'm done (yah, right!). Now I gotta stop reading this
forum so I can just enjoy the music.
Reply
post #1785 of 3981
11/4/08 at 8:18pm
Quote:
Originally Posted by funch View Post
Here are some new pix of my 'Blue' Bijou showing all the updates/mod's that
I've made since it was built.

http://i183.photobucket.com/albums/x...d/100_1521.jpg
http://i183.photobucket.com/albums/x...d/100_1517.jpg
http://i183.photobucket.com/albums/x...d/100_1520.jpg
http://i183.photobucket.com/albums/x...d/100_1516.jpg
http://i183.photobucket.com/albums/x...d/100_1518.jpg
http://i183.photobucket.com/albums/x...d/100_1509.jpg
http://i183.photobucket.com/albums/x...d/100_1510.jpg
http://i183.photobucket.com/albums/x...d/100_1511.jpg
http://i183.photobucket.com/albums/x...d/100_1513.jpg
http://i183.photobucket.com/albums/x...d/100_1514.jpg
http://i183.photobucket.com/albums/x...d/100_1515.jpg
http://i183.photobucket.com/albums/x...d/100_1519.jpg

Here's what I've done so far:
Added a separate heater tranny for the PS.
Regal's mod's.
Amphead's mod's.
Choke PS mod, variation 2.
6H30 gold pin input tubes.
ECC99 gold pin output tubes.
EZ81 PS rectifier.
Auricaps in C3, C4 and C6.
Changed signal wiring to cryo'ed Vampire (buried under all that shielding).
Stepped attenuator (50K) from ebay (Taiwan).

I just installed the stepper yesterday, and now understand the enthusiasm for them. The sum total of all the mod's has produced one fantastic listening experience. It's very detailed and dynamic, yet not at all fatiguing.

Thanks, Alex, for all your hard work on this design, and your continued efforts
in this thread. Thanks, also, to all of you who continue to tweak this great
design. I finally think I'm done (yah, right!). Now I gotta stop reading this
forum so I can just enjoy the music.
beautiful work!!!

i'm going with a blue/sapphire theme as well - i have some super bright blue LEDs to illuminate the interior of the amp.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Bijou All Tube Futterman Headphone Amplifier