PrimaLuna ProLogue Integrated - Help Needed

This is a followup for everyone: the working theory is that this is some sort of distortion-based feedback loop that happens apparently exclusively with the GoldenEar Triton speakers.

I took my PrimaLuna to a friend's house and tested with his B&W 80somethings. No matter how hard I pushed it, the pulsing issue did not occur.

I took the PrimaLuna to the dealer today and tried it with their floor model Triton Twos. I was able to reproduce the problem just as at home.

We hooked the amp up to some higher-end speakers (can't recall the model now) and could not reproduce the problem.

At this point the tech commented that I was simply pushing the amp too hard, and the Tritons were causing some sort of electronic or microphonic feedback. I don't entirely buy this explanation, but it does seem to be specific to the Tritons.

He then had the idea to replace the stock EL-34s with some KT-88s that he had on the floor.

Presto, problem solved. Problem no longer happens with the Tritons.

Again, I don't pretend to understand the issue, but apparently the EL-34s were freaked out about playing loud with the Triton Twos.

Thanks to everyone for their suggestions!

Thanks for the update. Glad you solved the problem.

I second both of Mofi's comments, Rob.

Your speakers are not especially hard to drive, as can be seen from John Atkinson's measurements of their impedance characteristics and sensitivity. Given that, and given that the problem occurred even when the speaker's built-in class D amplifiers were left unpowered (so that their fluctuating AC current draw could not influence the PL amp, perhaps by affecting the line voltage), what it all suggests to me is that the PL amp is marginally stable at very low frequencies when it is operating with essentially no load at very low frequencies. (The input impedance of the amps in the speakers, in combination with their associated low-pass crossover circuits, is undoubtedly vastly higher than the input impedance of a fully passive speaker, and hence represents a negligible load).

Whether or not an electrical feedback loop that is marginally stable at low frequencies will motorboat (i.e., oscillate at very low frequencies) could quite conceivably depend on minor variations in the gains, low frequency bandwidths, and other parameters of the parts and circuits that are in the loop, including the power tubes. And also on the setting of the volume control, if it is within the overall feedback loop or if the volume setting (and consequently the amount of power being supplied by the amp) has significant effects on internal operating voltages and/or the AC line voltage.

Given all of that, I wouldn't be surprised if the problem could be resolved even with EL-34s if you were to put a high-power resistor of say 30 to 50 ohms or so across the output terminals of each channel of the amp. That would provide the amp with a somewhat reasonable load at low frequencies, while drawing fairly little power at higher frequencies compared to the power draw of the speakers. A resistor of those values that is rated to handle 50 watts would provide plenty of margin in terms of power handling. A "non-inductive" resistor would be preferable.

The bottom line, though, if my analysis is correct, is that anyone contemplating using that PL amp with any speaker having a built-in low frequency amplifier (which would therefore present a very high impedance to the PL amp at low frequencies) should proceed with caution.

Best regards,
-- Al

Al,

Thanks for your additional detailed reply! I'm eager to try your resistor idea. Are we talking about something like this AG50 resistor? http://www.ohmite.com/cat/res_audiogold.pdf

I'm disappointed to report that the PrimaLuna is still showing the same issue with the swapped-in KT-88 tubes. The problem manifests in basically the same way, but the "motorboating" effect is much less noticeable (and the blue light "throbbing" is harder to see with these tubes), so I didn't see it right away.

I'm thinking I may have to abandon either the PL or the Tritons. I'm much enamored with the PL sound, so it might be the Tritons that get the boot.

Thank you again for your help.

Rob L.

Sorry to hear the problem is still there, Rob. Yes, the Ohmite Audio Gold resistors would be fine choices for this purpose, preferably in the non-inductive version that is indicated as being available. However, I couldn't find them indicated as being available anywhere in the 50W rating I suggested, and I couldn't readily find any other similarly high powered resistors that look suitable.

As you may have already seen, PartsConnexion carries the Audio Gold resistors in a 12 watt rating, in non-inductive form. What I'd suggest is that you purchase two (one for each channel) of the 47 ohm 12 watt 1% non-inductive resistors they carry (listed as "47R/12W, Ohmite-77999"), and on each channel connect one of them between the 4 ohm and common (0 ohm) tap of the amplifier (regardless of which tap you are using for the speakers). That shouldn't result in the resistor ever having to dissipate more than about 4 watts, well within its rating, regardless of what power tubes you may ever use in the amp and even with the amp running at max power continuously.

If you do that and it doesn't help, you could then try connecting the resistor between the 8 ohm and common terminals, which would require it to dissipate about 8 watts when the amp is putting out max power. That would also be reasonable as a long term solution if it solves the problem, although the margin relative to the 12 watt rating is a bit less than I'd ideally prefer to see.

Good luck. Regards,
-- Al