Bryston BHA-1 Measurements and Quick Impression

Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by purr1n, Sep 7, 2021.

  1. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I won't do much subjective reviewing here. The Bryston BHA-1 has been discussed much here. Subjective impressions can be found here: https://www.superbestaudiofriends.o...n-bha-1-amplifier-impressions-and-stuff.9450/

    It's also been mentioned as a good solid-state amp (fairly fast, zippy, without bad solid-state treble f'ery) that synergizes well with the HD800 here: https://www.superbestaudiofriends.org/index.php?threads/best-amps-for-hd800-2021-edition.10330/

    It has also received an SBAF Golden Schlong Honorable Mention (don't laugh - this is a serious award) here: https://www.superbestaudiofriends.org/index.php?threads/golden-schlong-awards-2020.10595/

    The only downside is that Bryston over the years has increased its price from just over a grand to $2295. For this, the BHA-1 also wins the SBAF Brown Starfish award.

    I will simply state a few things about the BHA-1 because there is a wealth of subjective impressions already. Is this amp the end all be all? Nope. I wouldn't buy it for the retail price. I think a lot of amps today may be competitive (but different sounding) at better prices today: iFi Audio Pro iCan, SPL Phonitor SE, Schiit Jotunheim 2, Violelectric HPA V280. Strengths of the BHA-1 are that one can largely discern the sound of different DACs through it. If we run a low-end AKM Velvet DAC, we will get murky mushy lows and mids and delta-sigma highs. If we run a hard hitting R2R DAC with smooth highs, we will get just that. This is when gear starts to get serious. The reason all DACs sound the same from Topping amps is because all DACs really do sound the same from those Topping amps, especially those with the TPA6120. Where the BHA-1 falls short is ultimate resolution and microdynamics. As such, the BHA-1 remains a fantastic deal at used prices and a good stop-gap if one is waiting for their ultimate TOTL tube amp to be built or even a good end-game if one wanted to stay with solid-state.

    Here is the standard 1kHz 0dBU (0.775Vrms) into a 300-ohm load. Inputs and outputs are balanced. Low gain (x5). The output of the analyzer was set to 0dBu. Because of the gain, the volume knob had to be turned down a bit to get 0dBu at the output.
    upload_2021-9-7_18-1-37.png

    I think I see a bit of second order with higher third and a bit of fifth order. It's hard to say because of the AC hum and harmonics. This is a bit disappointing because we are using balanced inputs and outputs. One thing balanced is supposed to do well is cancel stuff like this. I believe there was some talk about the BHA-1 not being a true balanced amp because the volume control isn't a four-gang pot - that there was a summing circuit somewhere. Regardless I wouldn't worry about this too much for this test case (Sennheiser HD6xx) since the AC noise is so far down. The highs look very clean though.

    Same deal here, except with SE output.
    upload_2021-9-7_18-10-34.png

    There are definitely more spurs in the highs from the SE output compared to the BAL output. I don't know if this means anything (I've only used the BHA-1 from its balanced outputs).

    FYI - don't compare these with ASR results. I do 0dBu output which is 0.775Vrms. That's enough to drive a nominal 300-ohm headphone such as the HD6** to 100db. I surmise ASR chose 2Vrms (and 4Vrms for BAL) because that's the level when the APx555 starts to differentiate itself from other analyzers. Below this voltage level, the performance is very close with other analyzers. Some people want to schlong their measurement gear. I don't feel this is necessary as these are just tools. In fact, I didn't even bother using the HPSA (high-performance sine analyzer) of the AP-555 with the above measurements.

    Furthermore, because the lowest gain is not unity, it's really not fair to compare to an amp that is measured under unity gain. x1 gain means more feedback = dull insipid sound. This is probably the reason why Bryston offered a minimum gain of x5. The downside of course is twitchy volume knob only usable near the most counterclockwise positions with efficient headphones.

    BTW, here are the approximate gain levels (from balanced input):

    SE output low gain: x2.5
    SE output high gain: x5
    BAL output low gain: x5
    BAL output high gain: x 10
     
    • Like Like x 17
    • Epic Epic x 3
    • List
    Last edited: Sep 7, 2021
  2. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Here is the GONAD panel. It's actually very good! Maybe too good, creeping into that territory. Yes there is something about too good (usually means high-gain stages with lots of feedback).

    upload_2021-9-7_18-48-40.png
     
    • Like Like x 8
    • Epic Epic x 4
    • List
    Last edited: Sep 7, 2021
  3. GoldenOne

    GoldenOne Friend

    Pyrate
    Joined:
    Jul 17, 2020
    Likes Received:
    1,049
    Trophy Points:
    93
    Location:
    England
    Nice stuff! And thanks for doing the measurements at a level which people are actually going to be using it at.
    Rather than 4v for..... Some reason
     
  4. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Volume / pot tracking
    R-L (Y) vs attenuation (X)
    upload_2021-9-7_19-19-1.png

    It's decent (~0.5db off) at most in the crucial -55 to -40db area. Well considering the high gain this amp has, maybe not so much - we could be using higher attenuation. Anyway, it's got one of those big blue alps. Told ya guys those Alps Blue Velvet deals really aren't that great. This is just the nature of pots.

    Will run power curves a bit later...
     
  5. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,467
    Trophy Points:
    113
    Location:
    Winnipeg
    This I will heartily agree on.

    The best amp I've used to tell dacs apart is the Benchmark AHB2 (driving my HE-6; and using monoblock AHB2's was only a tiny bit better). Since I sold the AHB2 and moved on to an SPL s800, I enjoyed the sound more and didn't feel like I was lacking in detail, yet the tiny smidges that help differentiate dacs was harder to extract. The BHA-1 to me fills that gap, even if it doesn't quite groove the way I like.
     
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    This is approximately 0.077Vrms into 30-ohms. Volume knob method was used (2Vrms input and volume adjusted to get close the output level). Amp set to low gain.

    upload_2021-9-8_8-23-7.png

    Efficient headphones in theory be fine. AC noise still there, but it's still quite far down at this very low voltage level. The main problem is barely any room on the volume knob. there is plenty of range until we hit the noise floor and AC noise. I don't think I even moved the knob an hour on the clock from full counterclockwise.
     
  7. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Checking THD across the spectrum for consistency. Bandwidth set to 90kHz. This is 0dBu (output set using volume knob) with 2Vrms input. Load is 300-ohms. Low gain. Note: this is not SINAD. It's THD. I'm stripping out the noise component.

    upload_2021-9-8_8-45-33.png
     
    Last edited: Sep 8, 2021
  8. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    THD (db) vs measured level (Vrms) . 300-ohm load. 1kHz
    upload_2021-9-8_8-57-39.png

    THD+N or inverse of SINAD (db). 300-ohm load. 1kHz
    upload_2021-9-8_8-59-9.png
     
  9. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    30-ohm load below.

    THD (db) vs measured level (Vrms). 30-ohm load. 1kHz
    upload_2021-9-8_9-4-28.png

    THD+N or inverse of SINAD (db). 30-ohm load. 1kHz
    upload_2021-9-8_9-3-5.png

    Homework / bonus points for the following:
    1. Calculate the max power level (30-ohms and 300-ohms) at 1% THD.
    2. Explain why THD+N / SINAD graph looks different from THD
    Yes, I expect you guys to think. I'm not ASR.
     
  10. GoldenOne

    GoldenOne Friend

    Pyrate
    Joined:
    Jul 17, 2020
    Likes Received:
    1,049
    Trophy Points:
    93
    Location:
    England
    1) 30 ohm: 1.77W
    300 ohm: (Little harder to tell exactly what voltage it is on the scale but I'm taking it as 24v so) 1.92W

    2) Cause THD (for most designs) will remain at a fixed % of the signal. Therefore as the signal lowers, it lowers by the same amount. Creating the 'flat' line on the graph.

    THD+N incorporates noise, and because the noise floor is fixed, it will be higher relative to the signal as the signal itself lowers. Creating the rising THD+N value as the signal level lowers
     
    Last edited: Sep 8, 2021
  11. Biodegraded

    Biodegraded Friend

    Pyrate Contributor
    Joined:
    May 28, 2017
    Likes Received:
    7,984
    Trophy Points:
    113
    Location:
    Vancouver BC
    For anybody remaining curious: how did he do ^ that? High school physics, specifically Ohm's Law:

    V = IR & P = VI
    V is voltage, I is current, R is resistance, P is power. Our dear leader has suggested tattooing these where you can see them.

    We have V & R so to solve for P, substitution => P = V^2/R.

    We're asked for the result at 1% (of the signal level) but we need that in dB to use the charts:

    dB level = 20log(1%) = -40, ie the result of interest is 40dB down from the 0dBu signal reference level.

    For the 300 Ohm load case, the curve reaches -40dB (y axis) at around 25V (x axis), so (25*25)/300 ~ 2W (GoldenOne used 24 for 1.92W)

    For the 30 Ohm case, -40dB is at about 7.25V so (7.25*7.25)/30 ~ 1.75W (evidently GoldenOne used 7.28V).

    Second question: yeah, the noise level remains the same regardless of signal level but distortion changes with signal level - so at low levels, THD+N is high (noise-dominated) and decreases linearly as level rises until distortion takes over from noise, whereas THD (no N) stays flat until distortion arrives.
     
  12. Tennstrong

    Tennstrong New

    Joined:
    Aug 31, 2020
    Likes Received:
    6
    Trophy Points:
    3
    Location:
    Vancouver, B.C.
    Thanks so much for writing this @purr1n! Been eyeing the BHA-1 for a number of months but it never seemed to get within the typically recommended price range for purchase (~900-1k, the last one around there iirc was on SBAF forums & prompted me to finally fold my lurking days & make an account here), and now like you mention they've increased the price further so the used market took a step on prices too.

    It's still definitely on my watchlist since it seems to really outlast a number of other amplifiers due to the lack of coloration while swapping other sound-chain pieces - I know you listed a number of amps that you feel provide a better valuation, but is there any in particular (listed or otherwise) you would consider to compete directly in this regard with the BHA-1? I'm guessing the list was something off the top of your head and not comprehensive with regard to the closest competitors, but if not would you say the headamp gsx mini would also be at that proficiency?

    Last - did I miss your phonitor SE impressions? I didn't see it in the phonitor thread & would be interested to hear your thoughts on that (well, if it is deserving of a written impression & not just the phonitor e with the xlr output cut).
     

Share This Page