[skeptic]

  Has anyone experimented with this amp with a 7236 or 6528 tube and used low impedance headphones like Grados? I am sure it's been done but it's hard to find comments from people that have done it.

 
For giggles, I tried some low ohm audio technica's and had a 5998 (same output imp. as 7236) in at the time.  The results were....not good, particularly in comparison to how great the amp sounds with high ohm phones.  If you want to drive low impedance phones with a bh kit, go with a s.e.x. or mainline.

 

[cspirou]

   
For giggles, I tried some low ohm audio technica's and had a 5998 (same output imp. as 7236) in at the time.  The results were....not good, particularly in comparison to how great the amp sounds with high ohm phones.  If you want to drive low impedance phones with a bh kit, go with a s.e.x. or mainline.

Thanks for the comment. I know that the standard response to this inquiry is to get the S.E.X. or Mainline, as is repeated on the BH forums. However there are many 6080 amps and some of those specify that they work down to 32 ohms. Given the DIY nature of the Crack and the fact that they encourage modding, there should be a way to mod the circuit to make 32 ohm headphones an acceptable option. If Woo Audio can do it with the WA3 and Pete Millett can do it with the Apex Teton, than it isn't a stretch to think that it's possible for the Crack to do this.
 
Or is any modification that accommodates low impedance headphones adversely affect the way the high impedance headphones sound and therefore is considered unacceptable?

 

[skeptic]

 
Thanks for the comment. I know that the standard response to this inquiry is to get the S.E.X. or Mainline, as is repeated on the BH forums. However there are many 6080 amps and some of those specify that they work down to 32 ohms. Given the DIY nature of the Crack and the fact that they encourage modding, there should be a way to mod the circuit to make 32 ohm headphones an acceptable option. If Woo Audio can do it with the WA3 and Pete Millett can do it with the Apex Teton, than it isn't a stretch to think that it's possible for the Crack to do this.
 
Or is any modification that accommodates low impedance headphones adversely affect the way the high impedance headphones sound and therefore is considered unacceptable?

 
The common method used to drop an OTL amp's output impedance is to apply negative global feedback to the circuit (or to add more tubes).  As I understand it (i.e. in overly simplistic terms), the problem with the feedback approach is that it does change the character of the sound and creates its own issues, like transient intermodulation distortion ("TIM").  (For reference, heavy use of negative feedback is how the o2 and similar chip amps hit their target measurements, but many (arguably most) ears don't ultimately find this sort of amp produces the best sound, when compared to other designs, because of TIM, possible oscillations and issues with disproportionate high order distortion.)
 
Bottlehead makes only non-feedback amps that avoid these issues and sound awesome when paired with the phones they are intended to drive.  The trade off is that you have to have go with a kit with output transformers, which add to the cost, if you want to drive low ohm phones.  
 
Also, FWIW, I owned the original WA3 for many years (sold after building and comparing to my crack) and can tell you that it did not play nicely (to my ears) with low ohm phones.  Maybe Woo has since added feedback to the circuit notwithstanding the downsides of doing so?  
 
Here's Pete Millet's explanation of what's going on in the Teton.  His comments about how the IEM circuit works are over my head.  
 

  The Teton has a highish output impedance similar to the original HA-2.  The Zout depends on what output tube you use...if I recall correctly it's around 100 ohms using a 6080, and as low as 50 ohms using a 6528.
 
Of course, the output impedance affects the sound, depending on what headphones you use.
 
I also added an "IEM" mode that trades gain for output impedance, so you can drive 16 ohm IEMs.  This is done with a simple resistive alternator, no magic.  So the gain is dropped by ~15dB, and Zout is lowered to ~10 ohms.
 
Although many would argue this amp is only good for higher impedance headphones (over 150 ohms), personally I find many lower-impedance phones (like Grados) sound better with a higher source impedance.  It depends on your taste.
 
Pete

 

[atomicbob]

 
Thanks for the comment. I know that the standard response to this inquiry is to get the S.E.X. or Mainline, as is repeated on the BH forums. However there are many 6080 amps and some of those specify that they work down to 32 ohms. Given the DIY nature of the Crack and the fact that they encourage modding, there should be a way to mod the circuit to make 32 ohm headphones an acceptable option. If Woo Audio can do it with the WA3 and Pete Millett can do it with the Apex Teton, than it isn't a stretch to think that it's possible for the Crack to do this.
 
Or is any modification that accommodates low impedance headphones adversely affect the way the high impedance headphones sound and therefore is considered unacceptable?

I have both the Crack and WA3+. Both use the 6080 as a cathode follower without NFB. Output circuit topology is identical. Component values differ a little. Those differences:
 
1. Voltage across 6080; WA3+ 119V, Crack 70V
2. Cathode resistor; WA3+ 1K5, Crack 3K
3. Output capacitor; WA3+ 470uF, Crack 100 uF
4. Output resistor shunt; WA3+ 10K, Crack 2K5
5. approx 33 ~ 35 mA current through the 6080 in both
 
I tried lower impedance (<100 ohms) headphones with the WA+ as well as the Crack. They did not perform their best in either. Lower impedance cans were better served by my V200 or ZDSE.

 

[GrindingThud]

Nice data/comparison. The higher mu tubes also help reduce output impedance quite a bit (5998/7236).
In my WA3, the PS sits around 190V depending on the tube. What I did notice was substantially different operating points in my amp when these biased up in the circuit...which I found interesting based on comments people make on how the different tubes "sound". The 6080 and 6N13S biased up around 38mA with 57V on the cathode resistor. 5998A and 7236 settled around 21mA with 32V on the cathode resistor. The 5998 is at the other extreme with only 18.6mA current and 28V on the cathode resistor. Impressions of tubes 'sound' is somewhat dependent on where on the curve its operating. I used the CCS approach to double the current in the 5998 to get it up higher on the curve....sounds better to me, and as many agree in this forum that speedball improves sound.
From a 32ohms perspective...gotta keep the output cap big (470 or bigger) to keep the low end frequency response from rolling off too much.

I have both the Crack and WA3+. Both use the 6080 as a cathode follower without NFB. Output circuit topology is identical. Component values differ a little. Those differences:

1. Voltage across 6080; WA3+ 119V, Crack 70V
2. Cathode resistor; WA3+ 1K5, Crack 3K
3. Output capacitor; WA3+ 470uF, Crack 100 uF
4. Output resistor shunt; WA3+ 10K, Crack 2K5
5. approx 33 ~ 35 mA current through the 6080 in both

I tried lower impedance (<100 ohms) headphones with the WA+ as well as the Crack. They did not perform their best in either. Lower impedance cans were better served by my V200 or ZDSE.

 

[cspirou]

Thanks everyone, thus is really good information. Although I have no clue what Pete Millet means by a 'resistive alternator'.

 

[atomicbob]

*edit*

What I did notice was substantially different operating points in my amp when these biased up in the circuit...which I found interesting based on comments people make on how the different tubes "sound". The 6080 and 6N13S biased up around 38mA with 57V on the cathode resistor. 5998A and 7236 settled around 21mA with 32V on the cathode resistor. The 5998 is at the other extreme with only 18.6mA current and 28V on the cathode resistor. Impressions of tubes 'sound' is somewhat dependent on where on the curve its operating.

*edit*

I've often wondered about this. If each of the tubes were biased in such a way as to put them into the same operating point on their respective curves, I wonder if the sonic differences would converge to something more similar than different? Conversely, re-bias the same tube into different operating points to achieve the audible variances described around the internet?

Do (or did before mods) you hear a residual hum with the WA3 if the room was very quiet and listening to say either HD600/800 or T1/T90 cans? I had a little in mine before I added an additional 10K resistor to the shunt such that the hum was reduced below my hearing threshold. It didn't go away as measured with the lab mic / preamp and measurement system. I often wonder if twisting the leads for the filament supply to the 6DJ8 would have resolved this.

 

[GrindingThud]

Yes, the amp has some hum that can be reduced by twisting the filament supplies and floating it with a pair of resistors split to ground, rather than grounding one side. No hum even with earbuds now. http://www.head-fi.org/t/94853/woo3-modified/225#post_9269260
I put a lot of the great ideas from the crack into my WA3.

I've often wondered about this. If each of the tubes were biased in such a
way as to put them into the same operating point on their respective curves, I wonder if the sonic differences would converge to something more similar than different? Conversely, re-bias the same tube into different operating points to achieve the audible variances described around the internet?


Do (or did before mods) you hear a residual hum with the WA3 if the room was very quiet and listening to say either HD600/800 or T1/T90 cans? I had a little in mine before I added an additional 10K resistor to the shunt such that the hum was reduced below my hearing threshold. It didn't go away as measured with the lab mic / preamp and measurement system. I often wonder if twisting the leads for the filament supply to the 6DJ8 would have resolved this.

 

[atomicbob]

Yes, the amp has some hum that can be reduced by twisting the filament supplies and floating it with a pair of resistors split to ground, rather than grounding one side. No hum even with earbuds now. http://www.head-fi.org/t/94853/woo3-modified/225#post_9269260
I put a lot of the great ideas from the crack into my WA3.

Excellent! thank-you!

 

[Jimmy24]

Finally got around to taking some pictures of my 7119's. 

 

 

 
12AU7/12BH7/5687/7119 Switches
 

 

[Guizan]

Going from the O2 to the Crack+Speedball would be benneficial in what ways? Or maybe what kind of differences can I expect? Never heard a tube amp (guitar amp, yes). I'd like to try something new, but I'm not sure if the law of diminishing returns is applicable here.

 

[atsq17]

Going from the O2 to the Crack+Speedball would be benneficial in what ways? Or maybe what kind of differences can I expect? Never heard a tube amp (guitar amp, yes). I'd like to try something new, but I'm not sure if the law of diminishing returns is applicable here.

I think you will be pleasantly surprised at how tangible the differences are assuming you have a high impedance headphone. I know several who have moved to Crack from o2 and it's like seeing the world in a small 2d screen to a large 3d one. Don't know if that's a good analogy but is very noticeable and i think the returns are very worthwhile.

 

[atsq17]

Very awesome. Does having a DAC dramatically give you a difference in sound? I'm still debating whether I should get one or just save my money for a new pair of headphones. I'm into getting a better soundstage so if a DAC can do that I might invest in one. How do the T90's sound in compared to HD 650's?

If you have a hd650 stick with it for now. Grab yourself an audio gd nfb 3 2014. Then enjoy the extra clarity details and Soundstage

 

[jinrum8]

  Going from the O2 to the Crack+Speedball would be benneficial in what ways? Or maybe what kind of differences can I expect? Never heard a tube amp (guitar amp, yes). I'd like to try something new, but I'm not sure if the law of diminishing returns is applicable here.

The review/comments here seems to give some comparison http://www.reddit.com/r/headphones/comments/1gkfaq/

 

[Jimmy24]

Atsq is right. Upgrading from the O2 to the Crack will really open up your headphones significantly. There's much more air between instruments, sound stage improved, and you'll hear details that you never heard from the O2. 

If you have a hd650 stick with it for now. Grab yourself an audio gd nfb 3 2014. Then enjoy the extra clarity details and Soundstage

 
I have not heard of that DAC before but will consider it when I purchase one. I mainly heard great reviews from the Schiit Bifrost for the Crack. Any noticeable differences between the two DAC's?