Joule Electra LA -150 ii/ 200 and impedance issues

Jud Barber says, in response to my e-mail, that his preamps are designed to work best with impedance 25K ohms and above. So I should be safe. thanks for your help, Al and Bifwynne

BTW, I'll mention another potential biamping issue that isn't generally realized.

In the case of a preamp driving a single power amplifier, if the output impedance of the preamp is high (at high frequencies), and the interconnect cable is long and also has high capacitance per unit length, the result can be a slight but perceptible rolloff of the upper treble. Although in most circumstances that effect won't be great enough to be audibly significant. But in the case of a preamp driving two power amps from a single output stage, the possibility of that kind of issue increases significantly, because the combined (total) capacitance of the interconnect cables to BOTH amps will affect the signal seen by the high frequency amp.

So in the situation where a single preamp output stage is driving two power amps, and the preamp has a highish output impedance, it becomes especially important to minimize the TOTAL of the lengths of BOTH interconnects, and/or their capacitance per unit length.

All of these considerations, btw, are also applicable to the situation where a preamp is driving both a power amp and a powered subwoofer.

Regards,
-- Al

So, before biamping, this calculation and these data should always be done and obtained, or you could end up with worse sound at great expense? WoW!?

Yes, unless the preamp provides two pairs of outputs that are driven by separate output stages (i.e., that are "individually buffered"). Some preamps do that, but most don't. With a preamp having that provision, the 10x guideline would be applied in the normal manner.

Interesting that, in all the various bi-amping posts, I've never come across this concept. Am I blind, is it less practically important or are others equally unaware of it?

It won't be an issue with most solid state preamps, due to the low output impedance they usually have. The high input impedance of most tube power amps will also reduce the likelihood of it being an issue, if both amps are tube-based. And if, as is common with tube preamps, their worst case output impedance within the audible frequency range occurs at 20 Hz (due to the impedance rise that occurs in the deep bass region as a result of the output coupling capacitor most tube preamps use), the problem may not be readily perceptible if the frequency range of the speakers (as modified by the room) or of the source material doesn't extend down to very low frequencies.

Regards,
-- Al

Interesting that, in all the various bi-amping posts, I've never come across this concept. Am I blind, is it less practically important or are others equally unaware of it?
You may have trouble answering this question:)

So, before biamping, this calculation and these data should always be done and obtained, or you could end up with worse soound at great expense? WoW!?

Springbok, yup it's just like Al said. Take a look at some of my posts which discussed how I loaded my ARC Ref 3/5 preamp's 2 outputs: Main 1 -- drives the main amp. Main 2: -- drives a single subwoofer. ARC recommends that the combined load impedance presented to the Ref 3/5 NOT be less than 20K Ohms.

Using Al's formula, I realized that I overloaded my preamp. In my case, I found a gent who custom designed a built a buffer device for me that presents Main 2 with a much higher impedance load. The device also (i) sums the left and right channels for the subwoofer without shorting the Mains and (ii) enables me to tap both loads into the preamp's balanced outputs.

It's a cool device and was NOT terribly expensive.

Thanks, Denis. You understand correctly.

If a single preamp output stage (for each channel) is driving two power amplifiers, either because the preamp has two pairs of output jacks that are jumpered together internally and driven from a single output stage, or if y-adapters are used to split a single pair of output jacks, the load impedance seen by that output stage will be considerably lower than either of the individual power amplifier input impedances.

As you indicated, if a preamp is driving a single power amp the ratio of power amp input impedance to preamp output impedance should ideally be 10:1 or more (at the frequency for which preamp output impedance is highest, and its value at some frequencies may be much higher than the specified nominal output impedance, especially in the case of a tube preamp). A lower ratio may or may not give good results, depending mainly on how the preamp's output impedance varies with frequency (the less variation the better).

If a single preamp output stage is driving two power amps the same principle applies, except that the load impedance that should be used to calculate the ratio is the parallel combination of the two amplifier input impedances. Which as I indicated will be lower than either individual input impedance, usually much lower.

As illustrated by the calculation in my previous post for your specific amplifiers, the combined parallel input impedance of two amplifiers equals the product (multiplication) of their individual input impedances divided by the sum of their individual input impedances.

Regards,
-- Al

Al, As always, your insights are extremely valuable and in the past have prevented me from bad decisions, like biamping on single terminals:)....but are you saying that, if a preamps outputs are paralleled, which yoy say is common, bi-amping may cause issues without our knowing it because the combined impedance load is too low for the preamp? Isnt it conventional wisdom that the ratio should be 10:1 of impedance output:input? Maybe this isnt often achieved and maybe bi-amping may adversely affect frequency response and, on the whole, be deleterious instead of additive? Do I misundertand you? This is surely a fundamental question?

Denis, I took a look at the PF review you referred to. The comment in the review about 100K, which I quote just below, makes no sense to me. Frankly, I don't think the author has any understanding of how gain and impedances inter-relate.

Matching it [the LA-200 line preamplifier] to an amplifier with impedance input more than 100,000 ohms is critical to obtain significant gain in the system. The circuit is set for 8 dB gain, but can be adjusted up to 12 dB. I left mine at the stock setting. Bear in mind that low impedance amps won’t be driven to their full output. This may or may not be important to you. But, it is a consequential characteristic of the LA-200 that limits its universality.

First, under any circumstances that are remotely within reason what may be affected to a significant degree by having too low a load impedance will be frequency response, and possibly phase response, not gain. Secondly, a blanket statement that "low impedance amps won't be driven to their full output" makes absolutely no sense, because different amps differ widely with respect to how much input voltage they require to be driven to full power.

The combined impedance of your 100K and 47K loads, when paralleled, is (47 x 100)/(47 + 100) = 32K.

Even if the output impedance of the LA-200 were as high as 32K, which would be utterly absurd, there would only be a 6 db reduction in overall gain as a result of the impedance incompatibility. For more reasonable output impedances, and the 32K load impedance, the gain reduction would be negligible. And although I haven't researched the gain and sensitivity of your particular power amps, 8 and 12 db line stage gains are in the right ballpark for most (although not all) applications these days, especially if digital sources are being used.

That said, the 150 MkII has an output impedance that is specified as being selectable to either 400 or 1200 ohms. Conceivably it is a good deal higher than those numbers at some frequencies, especially in the deep bass if its output is capacitively coupled. And as you indicated there doesn't seem to be an output impedance spec for the 200. If the two sets of output jacks that are provided on the 150 (and perhaps also on the 200) are driven from the same output stage, and if the worst case (highest) output impedance of that stage at any frequency is more than a small fraction of 32K, frequency response flatness (not gain) may be adversely affected.

I would suggest that you ask Joule what minimum load impedance is recommended for the two preamps, and if two pairs of individually buffered output jacks are provided. More often than not, when two pairs of output jacks are provided on a preamplifier they are driven from the same output stage, and are simply jumpered together inside the rear panel. In which case the single output stage (for each channel) would see the combined 32K load presented by your two amplifiers, rather than either of their individual input impedances.

Regards,
-- Al