Bi-amping With "Y" Splitter: Any Impedance Problem?
- Thread starter Ron Resnick
- Start date
No earplugs?! That explains it!
----------
I do the same (drive 2 amps) with my tube pre, but it uses a cathode follower output and has about 250 ohm output impedance and ~7 mA current capability. I can't imagine any SS pres having an issue driving a 67 KOhm load, even tube pres without a proper output stage would work.
Unless the two preamp outputs are separately buffered, the outputs are in simple parallel.
If the outputs are in simple parallel, then it doesn't matter whether you parallel the amps on the preamp side of the interconnect or you parallel the amps on the amplifier side of the interconnect, correct? Preamp outputs in simple, unbuffered parallel are just an internal "Y."
In both cases the preamp output stage is seeing the same lower impedance caused by the amps being in parallel.
If the foregoing is correct then why would these two connection configurations result in any difference in sound?
I guess because it's still the case that two runs of interconnects connected to the preamp output stage still result in a slightly different RCL load than one run of interconnect?
If the outputs are in simple parallel, then it doesn't matter whether you parallel the amps on the preamp side of the interconnect or you parallel the amps on the amplifier side of the interconnect, correct? Preamp outputs in simple, unbuffered parallel are just an internal "Y."
In both cases the preamp output stage is seeing the same lower impedance caused by the amps being in parallel.
If the foregoing is correct then why would these two connection configurations result in any difference in sound?
I guess because it's still the case that two runs of interconnects connected to the preamp output stage still result in a slightly different RCL load than one run of interconnect?
Are you saying there is a difference in sound? For starters, turn off the amps and mute the preamp, swap cables at either end, and see if the issue follows the cable or sticks with the amp.
If the preamp outputs are just parallel connections then the load seen by the preamp is basically the same no matter how you connect the amps. Assuming identical cables, then any differences could be from cable routing, i.e. one cable having more noise coupled into it.
If the amps are near each other, I would personally prefer to use one long high-quality cable from the preamp to the amps to minimize added capacitance and the chances of noise coupling into two different cables (plus it's cheaper to have one long cable), then split just before the amps and use short cables to the amps themselves. Your last statement is true; two 50' cables will double the capacitance of a single 50' cable. In practice, I would expect negligible (inaudible) differences from just that, but it is a difference.
As mentioned previously, gain and phase matching of the amps may (should) be (probably is much) more critical than the cable. In the past I had to tweak my crossover design to allow the best integration when I had a SS bass and tube mid/treble amp (plus learned ARC inverted phase on the 4-ohm vs. 8-ohm taps of my amplifier, fairly obvious by the deep "hole" at the crossover point when I got it wrong the first time).
HTH - Don
Last edited:
No, I have not compared these two configurations.
I'm afraid I don't know what issue you are referring to.
Yes, this is what I was suggesting -- this makes sense to me.
Thank you.
I was not sure from your post if you had tried and heard a difference already, or were asking in advance. Now I know, no worries!
NP.
^Agree^.
There is also more “loop room” for noise with parallel ICs.
It seems like could be a lot easier to get noise in.
Hence I would be running a single XLR type of IC microphone cable, and if it is going into RCA, then terminating that as needed on each end.
There's no loop though, unless the both amps don't do anything to isolate IEC ground from signal/chassis, then maybe the power cable grounds will complete a loop, but this is very unlikely.
You can run XLR but if it's terminated single ended you get none of the advantages of using XLR. It's actually worse as you have one unused leg in the cable which probably throws off the symmetry and reduces noise rejection.
If there is a long run, I would agree running one long cable and two shorter cables is a better solution as capacitance adds in parallel.
I foresee in my future the need to have a pair of monaural unbalanced (RCA in) to balanced (XLR out) transformers.
I found a company -- Pine Tree Audio -- which will make a custom device exactly for this purpose.
sites.google.com
Currently they make the device with a Jensen transformer. They are willing to make it with any transformer I specify. They probably would also use whichever brand of RCA and XLR connectors I specify.
Which transformer would you suggest? Might a Lundahl be more transparent than a Jensen?
I found a company -- Pine Tree Audio -- which will make a custom device exactly for this purpose.
Pine Tree Audio - SE2B Single-Ended to Balanced
SE2B Stereo Single-Ended to Balanced Converter
sites.google.com
Currently they make the device with a Jensen transformer. They are willing to make it with any transformer I specify. They probably would also use whichever brand of RCA and XLR connectors I specify.
Which transformer would you suggest? Might a Lundahl be more transparent than a Jensen?
Can't help, sorry Ron. Jensen has been around "forever" and is pretty much the standard for professional use and such, but I do not know anything about Lundahl. That said, Jensen has a range of transformers, and the last I used (some time ago) run around $1k each these days so not cheap (well, maybe cheap by WBF standards).I foresee in my future the need to have a pair of monaural unbalanced (RCA in) to balanced (XLR out) transformers.
I found a company -- Pine Tree Audio -- which will make a custom device exactly for this purpose.
Pine Tree Audio - SE2B Single-Ended to Balanced
SE2B Stereo Single-Ended to Balanced Converter
Currently they make the device with a Jensen transformer. They are willing to make it with any transformer I specify. They probably would also use whichever brand of RCA and XLR connectors I specify.
Which transformer would you suggest? Might a Lundahl be more transparent than a Jensen?
I would use Neutrik XLR connectors; they are pro-grade standard and very well regarded, though again there are no doubt audiophile connectors around. I don't have a suggestion for the RCAs; I would be laughed out for what I use.
As an aside, if your 50' run is single-ended, then I would use a quality cable with double-braid and foil for noise rejection, like the Belden cable I used in my simulation. I actually prefer Mogami quad-star cable for XLR runs that need good shielding and do not require flexibility for constant handling.
Thank you, Don.
I just got off the phone with Jesse of Pine Tree Audio. He's a very nice fellow who is very flexible and makes a variety of custom devices:
I just got off the phone with Jesse of Pine Tree Audio. He's a very nice fellow who is very flexible and makes a variety of custom devices:
Some nice-looking work! If you are going custom, you might ask about using mu-metal material (comes in hard or flexible rubber-like sheets) for magnetic shielding of the transformer. I did that for a MC step-up transformer case in the primordial past and it helped a great deal in isolating from TT motor noise.Thank you, Don.
I just got off the phone with Jesse of Pine Tree Audio. He's a very nice fellow who is very flexible and makes a variety of custom devices:
While I have fun experimenting with bi-amping, I think I would rather run single-ended interconnect to the natively single-ended Jadis, and use the transformer to send a balanced signal to the balanced-only input solid-state REI, rather than run balanced interconnect to the natively single-ended Jadis, and then use an XLR to RCA adapter on the Jadis.
Jadis is known for being a bit temperamental with noise, so I sort of assume better to make her happy?
Which scenario would you guess would be the better way to go? Why?
Thank you!
Is your preamp balanced or single ended?
I'm guessing se, if so I'd run a single RCA run to a location near the amps, then Y to the Jadis and use an adapter or a RCA>XLR cable with pin3 unused to the REI. The reason is you won't get the noise rejection of XLR using a single ended preamp anyways and the short run from the Y to the REI doesn't matter that much. The only thing a trafo would get you is galvanic isolation which probably doesn't matter, but there's a small chance it could.
If the REI really needs to have a balanced in, it's best done by adding a RCA input and trafo inside the amp. Or get an amp that has a RCA input.
I assembled some NCore amps for bass use and put a Jensen se>bal trafo in the amp as my tube gear is all se. They are very small and any reduction in audio quality due to the extra parts is not all that significant for bass use. If the REI has a bit of room in it, it wouldn't be difficult at all to add some RCA jacks and trafos, just don't use both RCA and XLR ins simultaneously without also using a switch.
I was thinking of higher frequencies like RF.
Hence “loop” versus “ground loop”.
They would be connected at the preamp, and then at the speaker.
The Nuetrik ProFi seem pretty good.
Mogami quad-star and Neutrik here as well.
And they also do XLR to RCA, and vice-versa.
There is not dishonour in taking a staged approach and wiring up the Jadis first, to see if it is noisy or not… before adding in the REL.
If single ended is noisy, then try the transformer back at the preamp end.
The Jenson ISO-Max is a “go to” .
If it does not sound good, then the REL will not help/fix it.
The Lara currently on audition is SE and balanced.
Jadis JPS2 is single-ended only.
The Jadis amps are single-ended only.
I would be perfectly happy with this scenario, but I have to check with Angus to see if that causes the REI a problem. The REI is true balanced from input to output, so I don't know if it will be happy with Pin 3 unused.
Thank you, but I won't be modifying any REIs.
Why did you choose a Jensen transformer for that purpose rather than a fancier, audiophile-beloved Lundahl?
Similar threads
