Switching preamps in and out

If the interconnections are RCA I don't think it would be a good idea even if the level controls on the power amp are turned all the way down. The problem with RCA connections is that during insertion the signal on the center pin is applied before the ground connection on the shell is made, and during removal the ground connection is removed before the center pin connection is broken. So there will be moments during insertion and removal when the zero volts that should be present on the center pin when no music is being played will not have a defined ground reference, and therefore may be seen as a very large and conceivably damaging voltage.

XLR connectors, on the other hand, make the ground connection first and break it last, so they do not have that issue. I suspect that it would be safe doing what you are proposing with XLRs, provided that you make absolutely certain to turn down the amp's level controls when changing connections. Personally, though, I still wouldn't do it.

Other approaches: If the output impedance of your source component is low enough in relation to the input impedances of the three preamps, you could use splitters or y-adapters to drive all three preamps at once, and a manual switchbox to select which preamp's outputs are routed to the power amp. If the impedance relationships in that situation would be unsatisfactory for good sonics, you could use a second switchbox to select which preamp the output of the source component is routed to.

If you want to consider those possibilities, and want further analysis of the impedance issue, post back with the specific models that are involved, and the preamp input impedances and source component output impedance if you know them.

The DB Systems DBP-2JAU/5 would be a suitable RCA switchbox. Click on the "photo" and "review" links for further info on it.

Regards,
-- Al

Hey, Wolfie, don't you think that is a bit of a bold statement, considering that you don't even know what components are being used, what the designs of their input stages are, what effect a brief overvoltage might have on their long-term reliability, what their grounding configuration is, whether or not they provide a ground lift switch, whether or not he's using cheater plugs on any of his components, whether or not any of his components have two-prong power plugs, and whether the connections are RCA or XLR?

Note also that I referred to a "CONCEIVABLY damaging voltage."

Have you ever connected the probe of an oscilloscope to a low or zero voltage circuit point on a piece of equipment, without connecting its ground lead, and looked at the voltage waveform that is displayed on the scope under that condition?

Regards,
-- Al

Mike, the output impedance of the MPS-5 is extremely low, as indicated in the third paragraph of JA's measurements in Stereophile, so I don't think that it would have any problem driving all three preamps at once, via splitters or y-adapters, even though the input impedance of the Allnic is only 10K. The input impedances of the other two preamps being much higher, 250K for the Messenger, and probably something like 67.5K for the VAC, which is what JA measured for the Mk II (non-A). The parallel combination of those three input impedances is 8.4K, which is hundreds of times higher than the output impedance of the MPS-5.

So IMO a good approach, assuming that the cost is worth it to you, would be to drive all of the preamps simultaneously, via splitters or y-adapters, and purchase the DB Systems switchbox to select which of their outputs is routed to the power amp.

FWIW, I used that switchbox extensively some years ago and I did not perceive it as having any sonic effects of its own when it was inserted into the system. Although my system then was not quite as resolving as it is now.

Regarding Wolfie's post, upon re-reading what preceded it I'm not sure that he was understanding that what I was talking about, given that the level controls on the amp would be turned down, was NOT the possibility of damage to the speakers (although I can envision ways in which even that might be possible). What I have been talking about is the possibility of either damage or degradation of long-term reliability of the input and/or output stages of the electronic components. If that wasn't clear, I hope he will re-read everything I said from that perspective.

Regards,
-- Al

Salectric, I would pose to you the question I posed to Wolf:

Have you ever connected the probe of an oscilloscope to a low or zero voltage circuit point on a piece of equipment, without connecting its ground lead, and looked at the voltage waveform that is displayed on the scope under that condition?

I assume that your answer, and his, would be "no." I have done that. Depending on the grounding configuration of the particular equipment, a 60 Hz sine wave measuring well upwards of 60 volts will sometimes be seen in that situation. And depending on the grounding configuration of the components in the audio system, a similarly huge sine wave might end up being applied to the inputs and/or outputs of all of the audio components that are involved in the interconnect swaps (including the source component), during the fraction of a second or so that the RCA plugs are being inserted or removed, due to the absence of a ground connection during that fraction of a second.

Do you really want to risk applying upwards of 60 volts to the inputs or outputs of components like these, at least some of which cost well over $10K, even for a fraction of a second? Inputs and outputs that are designed to generate or receive voltages that are around 2 volts or less? If so, be my guest. I would encourage others not to do so, even if the approach I suggested compromises the resolution of the comparison that will be made.

I am NOT saying that damage WILL occur as a result of the approach that you and Wolf are espousing. Most likely it will not. But I am saying that it creates a risk, especially given that RCA's are involved, and that it is therefore a very poor practice that should not be encouraged.

Regards,
-- Al

I'm going to put an email in to Kevin Hayes at VAC, and see what he thinks.

Mention to him that you are using the unbalanced RCA output of a source component that is solid state, for which the output impedance is 11 ohms.

I would expect that in general solid state components will have greater susceptibility to damage or degradation of long term reliability in this kind of situation than tube components.

Regards,
-- Al

I agree, Salectric, that the chances of there being a problem in this particular situation appear to be slim.

I would note, however, that my own VAC Renaissance 70/70 Mk III amplifier has a 3-position switch controlling its ground configuration. One of those positions isolates circuit ground from chassis ground (except at RF frequencies), and hence from the AC safety ground that as you indicate would otherwise prevent circuit ground from floating to voltage levels that are significantly different than those of the upstream components, when the IC's are disconnected. I don't know if any sort of ground lift provision is included in any of Mike's components.

I would also note that two-prong power plugs were not just creatures of the 1950’s and earlier. The STAX and Tandberg components in my system, from the mid and early 1980’s, respectively, all have two-prong power plugs. And some people have cheater plugs on some of their components.

I would also re-emphasize that the front end situation has to be carefully considered, as well as the preamp-to-amp interface. Especially given that the source component is solid state, per my comment to Mike just above.

One of my basic motivations in this thread has been to dispel the impression, which threatened to be created early on, that in a situation where no signal can be propagated through the power amp one should necessarily feel free to do anything and everything with interconnects while power is on. As you appear to realize, but others may not, there are a whole lot of variables that can make that assumption risky. Including those I mentioned in my initial response to Wolf:

… what components are being used, what the designs of their input stages are, what effect a brief overvoltage might have on their long-term reliability, what their grounding configuration is, whether or not they provide a ground lift switch, whether or not he's using cheater plugs on any of his components, whether or not any of his components have two-prong power plugs, and whether the connections are RCA or XLR….

Regards,
-- Al

Kevin's responsiveness never fails to amaze.

Enjoy!

-- Al