Balanced to single-ended converter (XLR to RCA / BAL to SE)

Discussion in 'DIY' started by purr1n, Jun 27, 2020.

  1. Elmer Danilovich

    Elmer Danilovich MOT:Earmen, HeadAmp, Bricasti; AKA:MShenay

    Contributor
    Joined:
    May 8, 2018
    Likes Received:
    346
    Trophy Points:
    63
    Location:
    In the Oak Forest
    My Jensen 1:1 Box shipped out today, and yes it is NOT the Pro to Consumer Box, so it'll be cool to see what you think when you get that solution

    An honestly I kinda don't quite need the converter any more tbh... so I'd be willing to send me around if we could get a little tour going with the @atomicbob Recommendation +XLR to RCA's & the 1:1 Box I got, just to see once and for all what works best without building it yourself [heck maybe some one can throw in their hand built box]
     
  2. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    The project with the WE111c turned out to be a little more time consuming than I had anticipated. These transformers are bulky and I decided to build two encolsures:
    [​IMG]

    I used step drill bit (this one) to make the Neutrik holes. I double checked with Schiit that the Gungnir A2 can drive a 600ohm load (the answer is yes), and wired up the transformers for 600:600. The transformers meausred 664 and 651 respectively when the secondary is loaded with a 600ohm resistor:

    [​IMG]

    Next up is installing everything in the enclosures and wiring up the input and output.

    Edit: I managed to put together the enclosures. Just need to wire it up.

    [​IMG]
     
    Last edited: Jul 26, 2020
  3. k4rstar

    k4rstar Britney fan club president

    Pyrate BWC
    Joined:
    Jun 11, 2016
    Likes Received:
    6,940
    Trophy Points:
    113
    Location:
    Toronto, Canada
    these are not cheap but sublime sounding line transformers. I currently use the much more affordable UTC A-20 for the purpose of stereo-to-mono mixing but I want to replace it with one of these eventually.

    your pair look very clean, I look forward to the finished product
     
  4. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    I bought the pair for $175 5 years ago. I wouldn't buy them at today's prices.
     
  5. kukur9

    kukur9 Acquaintance

    Joined:
    Aug 22, 2016
    Likes Received:
    84
    Trophy Points:
    18
    Location:
    Seattle, WA
    Anyone interested in a like-new Jensen Iso-Max PC-2XR? This is the 4dB (I think, I've forgotten step down) that I bought to try with a balanced DAC. Burned in a bit but ultimately didn't care for the change. Happy to just get rid of this so think "steep discount." Again, like new since I used it all of about 20 or 30 hours. Let me know.

    Also have a like new Emotiva CMX2 (Gen 2) that I tried but don't need. Also about 20 hours old.
     
  6. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    I finially got a chance to finish the converters this weekend:

    IMG_20200808_165433.jpg

    I didn't understand how gimped are the Gungnir Multibit's SE outs until I plugged in the converters (Gungnir A2/G5). I thought the difference was going to be subtle, but it was quite clear: the bass became more articulate and it 'lifted' (comparatively, the bass with the SE out sounded recessed). I recommend anyone relying on the SE outs to build a balanced-to-SE converter. I followed @purr1n 's diagram exactly (note how XLR-pin1 is not connected to the treansformer's ground tab, which is the left-most tab).
     
  7. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,461
    Trophy Points:
    113
    Location:
    Winnipeg
    Wow, when the output transformers are arguably bigger than your power supply iron...
     
  8. monacelli

    monacelli Friend

    Pyrate
    Joined:
    Jul 23, 2017
    Likes Received:
    1,649
    Trophy Points:
    93
    Location:
    The Land of Enchantment
    This is awesome. Hard not to be at least a little jealous of your system with those unobtainium Tannoys and these sweet transformers. Congrats dude, you're doing it right. Honestly, I think this is the way to go. If I had a DAC that couldn't drive 600-ohm loads, I would probably skip the transformer approach all together and use something active like Freya S instead. Sure input transformers are easier to drive, but you won't get the same bandwidth, treble extension, and bass clarity as with 600-ohm output transformers like yours. What's your impression of the overall tonality with the transformers in the chain? Did they affect Gungnir A2's cooler slightly analytical nature at all? I think it will be a really interesting question to see whether Nexus or some variant makes its way into Schiit's next gen output stages, at least for Yggdrasil. I have a hunch that if Gungnir and Yggdrasil were released today, there would not be such a noticeable disparity in performance between the balanced and single-ended outputs.
     
  9. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    First, totally jealous of @batriq’s Western Electrics. And his Tannoys. And his cool avatar. I’ve been researching vintage iron for a few weeks now and maybe I’ll be able to offer a comparable find from Telefunken that we could put head to head with the WE 111Cs.

    But, I’d like to take a moment to clarify a misunderstanding that has been repeated several times here on SBAF. Transformers, when connected between a source and a load, don’t present the amplifier with a fixed load (e.g., 600 ohms). Transformers simultaneously mirror the source and load impedances to the other, multiplied by the ratio of the turns squared.

    So, if you have a 47k ohm input impedance on your power amp, and a 1:1 “600 ohm” transformer, the load seen by your source is (1/1)^2 x 47k = 47k. Not 600 ohms! Any 1:1 transformer leaves the impedances unchanged!

    A 1:2 transformer multiplies the load impedance by (1/2)^2 = 1/4, so a 47kohm load looks like a 12k load. And the reverse is true: a 1 ohm source Z will now appear to the amplifier as a (2/1)^2 x 1 = 4 ohm source.

    Caveats: technically you need to add the DC resistance (DCR) of the windings, so if you calculate the effective load seen by the source, you need to add the DC resistance of the primary winding to that effective load, and vice versa. Usually the DCR is small compare to the true load and it can be neglected for a quick calculation.

    Also, if the transformer is unloaded, then it will present the source with some effective self Z, which is the DCR + inductance reactance (2 pi f L)—but this doesn’t matter when loaded, and this self Z is not why it’s called a 600ohm transformer.

    The 600:600 ohm nameplate rating (or 10K:10K, or whatever is specified) is the design target impedance—meaning these are (or more realistically, are vaguely close to) the impedances for the source and load required to achieve the frequency response the transformer designer intended (usually this is the maximum bandwidth point). Use a transformer with mismatched nameplate impedances (e.g. the 600:600 Western Electric on a modern amplifier) and it will work fine but you will get a different frequency response. This may be problematic or it may be desirable. And you can tune the extent of this effect by adding resistors across or in series with the coils, although at the cost of some SPL (because you’re creating a voltage divider).

    In sum, you don’t need a pro source to use a 600 ohm transformer. But you must be willing to pay the price in deviation from the design response. Why this happens relates to the second order filter effects of inductance and capacitance—I won’t get into calculating what to expect—but you can read about it in the below.

    References:
    https://www.eaton.com/ecm/groups/public/@pub/@electrical/documents/content/wp009001en.pdf

    https://www.sowter.co.uk/faq.php

    https://www.eevblog.com/forum/projects/transformer-impedance-and-dc-resistance/
     
    Last edited: Dec 3, 2020
  10. monacelli

    monacelli Friend

    Pyrate
    Joined:
    Jul 23, 2017
    Likes Received:
    1,649
    Trophy Points:
    93
    Location:
    The Land of Enchantment
    I suppose I hadn't thought about this carefully enough. Thanks for sharing these references. If this is the case, that a 600-ohm 1:1 transformer should be just as easy to drive as a 10k-ohm 1:1 transformer, then what can explain the difficulty of Yggdrasil A1 driving the 600-ohm input transformers in the EC Studio? It seems that some DACs are capable of this and others aren't. I was under the impression that this is one of the main things resolved by the A2 output stage. Here is the exchange where Marv shares the backstory between Schiit and Eddie Current:

    https://www.superbestaudiofriends.o...-multibit-impressions.158/page-42#post-243022
     
  11. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
  12. monacelli

    monacelli Friend

    Pyrate
    Joined:
    Jul 23, 2017
    Likes Received:
    1,649
    Trophy Points:
    93
    Location:
    The Land of Enchantment
    My hunch is that Craig put a compensation network in the Studio in order massage the load/source impedance to achieve the intended frequency response of the transformer (similar to the figure below). This compensation network (perhaps a simple resistor or two) could have presented the source with a low-impedance fixed load, most likely 600 ohms or close to it. This would obviously necessitate the output stage of the DAC being able to handle driving such impedances. So perhaps there isn't a difference between driving a 10k-ohm 1:1 transformer and an uncompensated 600-ohm 1:1 transformer, but to get the best possible performance out of a 600-ohm transformer, you may want to introduce a passive network that brings the source and load impedances more in line with the spec sheet. That's my best guess! @purr1n, what are your thoughts?

    Test circuit from the Jensen PO-2XX datasheet:
    https://www.jensen-transformers.com/wp-content/uploads/2014/08/PO-2XX.pdf
    [​IMG]
     
    Last edited: Aug 9, 2020
  13. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    Thank you for the kinds words!

    In the main system with the Tannoys, I'm using a Yggdrasil A2 and Freya S. The Gungnir is in a test system I've been using to try out different setups (e.g. mostly amps I refurbished over time, and various sources). But now with the Zu's moved to that system plus the amp coupled to the Gungnir with these transformers, it's sounding really good (I don't feel like changing much if anything in it).

    Yes, the Gungnir no longer sounded too cool. The primary difference I heard is in the presence of the bass/upper bass, which may be why.
     
    Last edited: Aug 9, 2020
  14. Melvillian

    Melvillian Friend

    Pyrate
    Joined:
    Jul 3, 2017
    Likes Received:
    1,345
    Trophy Points:
    93
    Location:
    Los Angeles
    These sound really good. Compared to the single ended out on the Gungnir A2, it's cleaner, more width and depth, more dynamic, better bass slam and texture. It's just better all around and doesn't have the smoothed over and warm sound of the Jensen 4:1.

    I wanted to have the XLRs on the left and RCAs on the right but this layout seems to work better. Would have preferred a smaller box but this will do. It’s almost the same size as a bifrost. Orange, black, grey, and white are all grounded to the chassis. Seems to be working well with no hum. Not sure if I should disconnect the grey altogether or

    edit* adding a link to the documentation from cinemag, which is where the diagram is from
    https://cinemag.biz/line_input/PDF/CMLI-15-15B.pdf

    [​IMG]
    [​IMG][​IMG]
     
    Last edited: Aug 23, 2020
  15. Baten

    Baten Friend

    Pyrate
    Joined:
    Mar 18, 2018
    Likes Received:
    1,130
    Trophy Points:
    93
    Location:
    EU
    Looks funky! what converter is that?
     
  16. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    I believe it’s a Siemens preamp circa 1981 for an analog mixing console. The transformers are what I’m after. They have a good reputation and reportedly extremely high bandwidth when used with consumer-gear loads. I hope so because I bought a bunch of them.
     
  17. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    View attachment 22829
    Holy cow! With the Haufe RK283, the sound stage got huge and the imaging is more disconnected from the speakers. The sound is also more detailed relative to the Adagio’s built-in Lundahls.

    This is NOT a small difference. It’s as big as a difference as a good vs. midrange DAC. The single-ended use of Adagio has really hidden its true potential. And while I had hoped for more detail, the improved stage was a real surprise.

    Looking forward the the Cinemags.
     
    Last edited: Sep 5, 2020
  18. dBel84

    dBel84 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,229
    Trophy Points:
    113
    I needed a conversion box to experiment with DSP and picked up a sescom SB-1 mkii balanced box buffer.

    [​IMG]

    [​IMG]

    It does a whole load of things a pro mixing desk might need but didn't really work as I had hoped.

    So I went active

    [​IMG]

    [​IMG]

    Basically 2 different approaches to allow either SE to Balanced or Balanced to SE

    Thanks @MisterRogers for sharing his BAL/SE boards, I used the same approach as Alex Cavalli used and made up boards for the SE/BAL

    It works perfectly and I am hard pressed to say I can identify the additional (nasty opamps) in the signal chain.

    Nowhere near as sexy as large transformers but we do what we must.

    .. dB
     
  19. gixxerwimp

    gixxerwimp Professional tricycle rider

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    5,773
    Trophy Points:
    113
    Location:
    small island claimed by China
    @dBel84 for us dummies, can you explain a bit more about what's in your box(es)?
     
  20. dBel84

    dBel84 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,229
    Trophy Points:
    113
    The pro sescom unit specs are below

    [​IMG]

    The balanced to SE is passive using a transformer - not sure what it is . Looking at the circuit , it looks like it sums the input through a resistor network and then uses the transformer 1:1 . The SE to balanced has gain and uses an opamp to drive the transformer input

    - I found the effect of this unit to significantly reduce the signal volume and when using the gain for SE it audibly distorted past 12 o'clock

    The DIY effort :

    SE-BAL : This uses a buffer in front of the DRV134/135 chip which does all the work = bigger green board

    BAL-SE : this uses opamps to sum the balanced signals and then drives a single opamp unity gain output ( very similar but reverse of the inner workings of the DRV134 ) = smaller black board.

    the rest is just a simple dual 15V supply using standard voltage regs .

    ..dB
     

Share This Page