Crack w/ 600ohm headphones

Dr. Toobz · 3494

0 Members and 1 Guest are viewing this topic.

Offline Dr. Toobz

  • Sr. Member
  • ****
    • Posts: 432
on: January 27, 2013, 07:11:16 AM
I recently downsized my headphone collection to a single high-impedance unit, the DT880 600 ohm variant, which will only be used with my Crack amp. Given that I don't plan to use anything else (for the time being) with the amp, I was thinking I could decrease the output coupling cap value a bit and still maintain the same -3dB rolloff point in the low frequencies as I would have experienced into my 300 ohm HD650's (e.g., below 10Hz). The smaller cap size will give me a bigger budget for something exotic, as well as allow for easier and more stable mounting to the chassis plate. How low of a value in uF should I go?

I previously had put some electrolytic capacitors in the Crack to "wake up" my Senns, but found them to be rather grainy and "zingy" sounding once I switched to the Beyers. So, I'm back to film!



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19750
Reply #1 on: January 27, 2013, 07:23:32 AM
Hello Angelo,

Here is a convenient calculator that you can try;

http://www.v-cap.com/coupling-capacitor-calculator.php

The headphones and loading resistors present an impedance of around 500 Ohms, and 5Hz is a decent -3dB target, leaving you with a resulting cap value of 63uF (with the 100uF caps, you'd be at 3Hz).

Now, if you're going to decrease the output coupling cap, you can increase the 2.49K loading resistors a bit.  If you're going to have about half the capacitance, then let's push those up to 4.7K resistors, which gives you a 530 Ohm load (yay!). This gets you down to 60uF, though I suspect anything from the mid 50uF-70uF range is going to work well.

Do note that increasing the value of those 2.49K resistors without decreasing the size of the coupling capacitors is a bad idea.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Dr. Toobz

  • Sr. Member
  • ****
    • Posts: 432
Reply #2 on: January 27, 2013, 07:31:26 AM
Awesome - that's very helpful information! 60uF definitely puts some choices into play that are either too expensive or too ungainly in size (or both!) when forced to use 100uF. I even had gone to a 7802 output tube and was looking into even bigger capacitor values at one point, but I have determined that none of the sub-100 ohm headphones I have owned (or tried) were worth the effort.

Loving my new DT880's, by the way. There definitely is a difference between these and the earlier, 250 ohm "pro" version I owned when first building the Crack back in April, 2010. The 600's are more balanced and "effortless" sounding and not nearly as strident in the highs, and this is only with 10 hours or so on the odometer....



Offline Grainger49

  • Hero Member
  • *****
    • Posts: 7175
Reply #3 on: January 27, 2013, 08:31:04 AM
I wonder, just how low does most music go?  -3dB at 50 Hz is probably identical to -3dB at 30 Hz with 95% of music. 

It might be worth testing with cheap caps first.
« Last Edit: June 10, 2014, 07:48:24 AM by Grainger49 »



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19750
Reply #4 on: January 27, 2013, 08:35:30 AM
Yeah, that is very true, but people tend to be very fussy about that without ever listening and evaluating themselves.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Dr. Toobz

  • Sr. Member
  • ****
    • Posts: 432
Reply #5 on: January 27, 2013, 09:03:12 AM
I wonder, just how low does most music go?  -3dB at 50 Hz is probably identical to -3dB at 30 Hz with 95% of music.  

It might be worth testing with cheap caps first.

In my case, probably not very low at all. Most of my music is jazz, so the lowest tone would be an open E on the string bass, which is 41Hz, and that's not used terribly often! 40 or 50Hz is probably ample. The only thing I have noticed is that a higher rolloff point sometimes leads to a "bass bump" in the midbass frequencies with some headphones, which can make things sound muddy. However, this has only been when using transformers that roll off too high.




Online Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5833
Reply #6 on: January 27, 2013, 09:31:24 AM
PB's suggestion of experimenting is very worthwhile. Impulsive sounds - thump of a bass string or drum - have some spectral components down to DC, unlike a steady-state sine wave. If you are going to spend big bucks, take some of that budget to buy cheap caps - say, 10-22-33uF - and see where the point of diminishing returns actually is with your cans. This is how the BeePre caps were selected. You can use the leftovers to bypass the PSU caps, perhaps!

Paul Joppa


Offline Doc B.

  • Administrator
  • Hero Member
  • *****
    • Posts: 9657
    • Bottlehead
Reply #7 on: January 27, 2013, 10:44:08 AM
I will point out that we didn't just choose the Bee Pre cap value by ear. We did measurements, finding the point where a bigger value didn't seem to make any difference into about a 10K load.

Dan "Doc B." Schmalle
President For Life
Bottlehead Corp.