BottleHead Crack technical measurements

Discussion in 'Headphone Amplifier Measurements' started by atomicbob, Nov 29, 2017.

  1. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    Bottlehead Crack technical measurements

    If you are unfamiliar with audio measurements please use a search engine with the query:
    "audio measurements" or "audio measurement handbook"
    Look for publications by Richard C. Cabot and also by Bob Metzler, both from Audio Precision. There are other useful publications as well. These will provide basic knowledge.
    Interpretation of the following measurements is beyond the scope of this post.

    The data presented were collected as follows:

    1. PrismSound dScope III, picoscope 5243B
    2. 300 ohm load used for measurements
    3. Unbalanced cables Worlds Best Cable Gotham GAC-2 RCA
    4. Vaunix Lab Brick USB hub for measurement equipment
    5. 0 dB gain @ 0.775 Vrms 0 dBu except as noted, typically -16 dBu (0.122 Vrms)
    6. Bottlehead Crack gain knob at approximately 10:30 to achieve 0 dB gain
    7. Nearly stock Crack:
    a. No Speedball
    b. film bypass caps added to output electrolytic capacitors
    c. rebiased for e80cc instead of 12au7
    d. Tungsram e80cc driver
    e. Philips JAN 6080WC output

    Reference:

    0.775 Vrms @ 30 ohms is 20 mW producing approximately 101 dB SPL in HE-500 with sensitivity of 89 dB/mW
    0.775 Vrms @ 300 ohms is 2 mW producing approximately 101 dB SPL in HD650 with sensitivity of 98 dB/mW
    0.122 Vrms @ 300 ohms is 50 uW producing approximately 84 dB SPL in HD800 with sensitivity of 97 dB/mW


    Highlights:

    The BH Crack is a simple audio path and provides a great auditory experience with HD6x0 and HD800 headphones.
    THD and THD+N is consistent across the spectrum and is 2nd harmonic dominant. 3rd harmonic is 40 dB lower than 2nd.
    I still enjoy numerous evenings listening to this amp. Tube rolling enables many addictive sonic signatures.


    Picture of the Audio Lab setup for Bottlehead Crack / Liquid Crimson measurements
    DSC_0005a.jpg

    300 ohm load part 1 of 3

    Bottlehead Crack square wave 20 Hz 2 Vpp 10 mS / div 300R load
    20171128 SigGen Crack 20 Hz square 2000mVpp 10mS div 100KHz filter 300R.png

    Bottlehead Crack square wave 20 Hz 2 Vpp 10 uS / div 300R load
    20171128 SigGen Crack 20 Hz square 2000mVpp 10uS div 100KHz filter 300R.png

    Bottlehead Crack square wave 20 Hz 2 Vpp 2 uS / div 300R load
    20171128 SigGen Crack 20 Hz square 2000mVpp 2uS div 5MHz filter 300R.png

    Bottlehead Crack amplitude - phase - gain (A01) 300R load
    20171128 Crack A01 amplitude - phase - gain 300R.png

    Bottlehead Crack signal to noise (A07) 300R load
    20171128 Crack A07 SNR 300R.png

    Bottlehead Crack THD+N THD nth harmonic distortion (A04) 300R load
    20171128 Crack A04 THD+N  THD  nth-HD 300R.png

    Bottlehead Crack THD+N THD nth harmonic distortion (A04) with FFT 300R load
    20171128 Crack A04 THD+N  THD  nth-HD FFT 300R.png

    Bottlehead Crack Crosstalk (A08 A09) 300R load
    20171128 Crack A08 A09 Crosstalk 300R.png


     
    Last edited: Jan 27, 2018
  2. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    300 ohm load part 2 of 3

    editorial - I believe there is a tendency to focus too much on numbers such as THD attempting to reduce a complex topic to a single number representing a sort of goodness rating. Graphs can present far more data conveying a better understanding of the measurement over a range of frequency, levels, etc.

    Here are graphs for the various measurements

    Bottlehead Crack Frequency Response and Interchannel Phase pg 1/2 300R load
    G2_L1T2_1_A.png

    Bottlehead Crack Frequency Response and Interchannel Phase pg 2/2 300R load
    G2_L1T2_1_B.png

    Bottlehead Crack Gain vs Amplitude Left 300R load
    G2_L1T3_1_A.png

    Bottlehead Crack Gain vs Amplitude Right 300R load
    G2_L1T3_1_B.png

    Bottlehead Crack THD+N vs Frequency Left 300R load
    G2_L1T5_1_A.png

    Bottlehead Crack THD+N vs Frequency Right 300R load
    G2_L1T5_1_B.png

    Bottlehead Crack THD+N vs Amplitude Left 300R load
    G2_L1T6_1_A.png

    Bottlehead Crack THD+N vs Amplitude Right 300R load
    G2_L1T6_1_B.png

    Bottlehead Crack Crosstalk Left to Right vs Frequency 300R load
    G1_L1T10_1.png

    Bottlehead Crack Crosstalk Right to Left vs Frequency 300R load
    G0_L1T11_1.png

     
    Last edited: Nov 29, 2017
  3. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    300 ohm load part 3 of 3

    Bottlehead Crack THD THD+N vs Frequency with 1 KHz spectrum Left and Right 300R load
    20171128 Crack THD THD+N vs Freq 300R.png

    Bottlehead Crack 1 KHz spectrum Left and Right 300R load
    20171128 Crack 1 KHz THD+N THD FFT 0dBu 300R.png

    Bottlehead Crack 50 Hz + 7000 Hz spectrum Left and Right 300R load
    20171128 Crack 50+7000Hz 0dBu 300R.png

    Bottlehead Crack 600 Hz + 1700 Hz spectrum Left and Right 300R load
    20171128 Crack 600+1700Hz 0dBu 300R.png

    Bottlehead Crack IMD spectrum Left and Right 300R load
    20171128 Crack 19+20KHz -16dBu 300R.png

    Bottlehead Crack residual noise spectrum Left and Right 300R load
    20171128 Crack residual noise A=0dB 300R.png

     
    Last edited: Nov 29, 2017
  4. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    Residual noise on the Bottlehead Crack is quite low. Most people will never perceive the slight hum I detect when listening in my NC15 acoustic lab (unnaturally quiet as @Torq put it.) I decided to modify the power supply filter and lower the residual hum below masking threshold in my listening space. Blackground is much improved, eliminating the last niggling annoyance given how much I enjoy this amp's euphonic presentation.

    Power supply filter in the Crack is C-R-C-R-C with C=220uF and R=270 ohms. I replaced the first R with a Triad C7X choke, L=10H, R=270 ohms. Also bypassed the final C with a 1uF film capacitor.


    For easy comparison here is the result shown in a post above prior to modification:

    Bottlehead Crack THD THD+N vs Frequency with 1 KHz spectrum 300R load before PS choke mod

    20171128 Crack THD THD+N vs Freq 300R.png

    Below are the modified results. Compare these with the respective data presented in prior posts. The big winner was a reduction of the 120 Hz mains noise component by 10 dB. Also note a small 42 KHz spike was eliminated (compare 1 KHz spectrum graph to the unmodified PS 1 KHz spectrum graph in post #3 above). THD and THD+N were lowered a small amount. The residual noise spectrum overall was cleaned up a bit.

    Bottlehead Crack THD THD+N vs Frequency with 1 KHz spectrum 300R load after PS choke mod
    20171222 Crack THD THD+N vs Freq 300R - PS choke mod.png

    Bottlehead Crack 1 KHz spectrum Left and Right 300R load after PS choke mod
    20171222 Crack 1 KHz THD+N THD FFT 0dBu 300R - PS choke mod.png

    Bottlehead Crack 50 Hz + 7000 Hz spectrum Left and Right 300R load after PS choke mod
    20171222 Crack 50+7000Hz 0dBu 300R - PS choke mod.png

    Bottlehead Crack 600 Hz + 1700 Hz spectrum Left and Right 300R load after PS choke mod
    20171222 Crack 600+1700Hz 0dBu 300R - PS choke mod.png

    Bottlehead Crack IMD spectrum Left and Right 300R load after PS choke mod
    20171222 Crack 19+20KHz -16dBu 300R - PS choke mod.png

    Bottlehead Crack THD+N THD nth harmonic distortion (A04) with FFT 300R load after PS choke mod
    20171222 Crack A04 THD+N  THD  nth-HD FFT 300R - PS choke mod.png


     
    Last edited: Dec 23, 2017
  5. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    Great job as usual! I believe this is the first time I've seen an all tube amp measurement from you and glad to see the Crack even though OJ posted his own measurements before.
     
  6. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    Updated post 4 with data demonstrating improvements from power supply modifications.
     
  7. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,833
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    What tubes are these tests?

    Can you do a few different tubes to show the FR change?

    I am quite actually surprised at the bad crosstalk measurements. Maybe that's why is sounds so much like vinyl.

    And a yes, very euphoric.
     
  8. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    He's not using the Speedball upgrade with the CCS loads so not too suprising. Not only does the CCS limit power supply ripple, but also reduces the signal that comes from the other channel since the Crack uses a single unregulated power supply. This is a major reason why people use dual-mono power supplies.
     
  9. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,833
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    I knew it was sans Speedball, but still...I mean vinyl has mono bass from like 250Hz and above. I wonder what it would look like by doing the polarity reversal trick - inverting one channel on input and inverting back,on output. It would be interesting to see if this did anything to the channel separation.
     
  10. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    Don't you mean 250Hz and below. I don't see what polarity reversal would do.

    Crosstalk is one of those features that I'm okay with on headphone amps but I would not want it on speaker amps. Headphones give an unnatural presentation and a little bit of the signal bleeding over to the other channel is fine.
     
  11. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,833
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    Yes I did.

    This: http://superbestaudiofriends.org/in...-single-channel-polarity-phase-reversal.2942/

    I would love to see measurements of this as I would think it would only benefit crosstalk.

    To me it kind of depends on the amp and in what FR bands we are talking, which is why I appreciate these graphs. It can have a kind of crossfeed effect. I'm all for accuracy and if you want crossfeed then it can be added.

    Incidentally, the Valhalla 2 also has CSS and it only lists crosstalk at -71db. Not that that spec tells much as it sounds fantastic and still way above vinyl and FM radio, which I see is about anywhere from 35-25db. FM radio sometimes 20-15db.
     
  12. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    I updated post #1 with tube information.
    Tungsram e80cc and Philips JAN 6080WC as pictured
    Not likely to return to the BH Crack for measurements for awhile. I am 4 posts behind on devices measured but not yet posted. There are at least another 8 to 10 devices ahead in the non-day-job queue. High quality measurements take time for setup, thermal equilibration, actual measurements, documentation and posting. In one of the measurement threads I outlined why it takes at least half a day, even for just a few measurements.

    Correct. I have purposely not installed the Speedball in this BH Crack to exploit some of which you describe. There are plenty of other amps in the lab inventory with nearly dual mono supplies and other attributes if I wish to listen to the presentations they deliver. My BH Crack has been tailored for the specific sound it is delivering with this particular configuration, which I find especially appealing with 50s ~ 60s jazz and 60s ~ 70s rock.
     
  13. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    In the context of what Torq was talking about, he meant maximizing the power draw between the positive and negative voltage rails. Crack only has a single voltage rail and you wouldn't really see a difference anyway because crosstalk measurements involve keeping one channel silent and measuring the silent channel while the other one is active, phase wouldn't affect this. Plus this is a single ended Class A amplifier, the power draw is constant regardless of whether any music plays or not.

    As for accuracy, this is a headphone amp. Except for binaural audio, everything is mixed for speakers.
     
    Last edited: Jan 27, 2018
  14. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,833
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    Maximum power draw only benefits the performance of the amp. I was only asking could this possibly influence crosstalk as well. Only measurements can decide this. This is why I asked.

    And I know how Class A amps work. I have built some. Single ended and push-pull. Thanks.

    Totally understandable. And it is what it is. Totally for 50s ~ 60s jazz and 60s ~ 70s rock. Reminds me of those old tube amps from that era (HH Scott, The Fisher, etc.)
     

Share This Page