Bi-amping With "Y" Splitter: Any Impedance Problem?

[DonH50]

One amplifier has an input impedance of 200,000 ohms, and the other amplifier has an input impedance of 100,000 ohms. So I think the calculated input impedance seen by the preamp is 66,000 ohms.

The interconnect run is 50 feet long. I am using low capacitance cable.

With the Y splitter near the end, so short cables to the two amplifiers? I started to run the numbers but am lazy. I don't think it will be enough to matter. Here is a simulation for 50' of typical single-ended RG-6 type cable at 16.2 pF/ft, 1 k-ohm preamp output, 66,667-ohm load in parallel with 2 nF for the amplifiers, ignoring the splitter (assumed short so insignificant contribution. The RG-6 model includes distributed resistance, capacitance, and inductance. Loss at 10 Hz is 0.13 dB, and the -3 dB bandwidth is 58.4 kHz. Doubt it will be a problem for you.

You may have to watch out for a ground loop using a single-ended run.

 

[bonzo75]

8w 300b biamped with class D amp. Will be good to hear Ron's videos when he finishes biamping

 

[Ron Resnick]

is this a passive preamp?

no, active preamp

 

[Ron Resnick]

With the Y splitter near the end, so short cables to the two amplifiers? I started to run the numbers but am lazy. I don't think it will be enough to matter. Here is a simulation for 50' of typical single-ended RG-6 type cable at 16.2 pF/ft, 1 k-ohm preamp output, 66,667-ohm load in parallel with 2 nF for the amplifiers, ignoring the splitter (assumed short so insignificant contribution. The RG-6 model includes distributed resistance, capacitance, and inductance. Loss at DC is 0.13 dB, and the -3 dB bandwidth is 58.4 kHz. Doubt it will be a problem for you.

You may have to watch out for a ground loop using a single-ended run.

View attachment 135055View attachment 135056

Thank you very much for this analysis!

 

[DonH50]

no, active preamp

SS or tube? 1 k-ohm is pretty high for a typical preamp, though might be reasonable for a tube pre-amp. Whatever, the engineering side says it should work fine. I make no prediction about the audiophile part.

 

[Ron Resnick]

SS or tube?

tube

the engineering side says it should work fine. I make no prediction about the audiophile part.

Thank you.

 

[DonH50]

Somehow forgot to post, or deleted, the schematic for the previous simulation. This uses LTSpice, a free Spice simulator that is pretty well-regarded and easy to use.

IL = length = 50'
Rc = center resistance/foot, Lc = inductance/foot, Cc = capacitance/foot, RSc = shield resistance/ft
Rpre = preamp output = 1 k-ohm, Ramp = amp input = 66.667 k-ohm (both in parallel), Camp = amp input = 2 nF

The coax is Belden RG-6 type 1530A, a common coax for audio/video systems. O1 is a lossy transmission line model used to simulate distributed RLC parameters.

 

[Audiohertz2]

No he is trying to keep Jadis for midrange and have the Westminster for power.

Interesting , so how does Ron plan on gain matching , not to mention the odd load to the preamp , the different Sonic hand offs will be noticeable with the Clarysis bass panel running as high as 500 hz..!

IMO , It’s best to run an active xover with gain stage in this situation , passive parallel is best with identical amps for load and gain.

Ron also need to verify if both amps are non inverting, same phase is important ..

Good way to play around and burn sometime i guess ..