Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
In Reply to: RE: Improved Marantz 9?? posted by tube wrangler on March 21, 2017 at 18:14:38
The Marantz 9 runs the tubes conservatively at 55 ma with 440V on the plates. The output tubes last a long time on it. I have been running two quads of NOS Mullards and the getter flashes look nice and shiny after 3 years of use, averaging a few hours a day.
The amplifier also runs very cool. After 3 hours the transformers are barely warm to the touch. All components run as conservatively as possible.
As for instability, it is rock stable needing only occasional bias tweaking. Remarkable considering the push pull parallel design.
The only minor criticism is the slightly over elaborate input stage. But this was done to facilitate connecting a phantom center channel speaker, which was a feature used back in the early 60s to combat the so called "hole in the middle" caused by the speakers of the day. An added side effect of this is the ability to invert the phase or polarity of the signal, which can come in handy at times...
I think most folks disparaging the amp, gave it away for peanuts back in the 70's and now are kicking themselves.
Or folks who cannot afford a pair...
55 mA. at 440 VDC P-K ( amp uses fixed bias as I recall ) is 24.2 Watts dissipation of the EL-34. Looking up Mullard data sheets, they list 25 Watts as the EL-34's Maximum plate dissipation, in pentode mode.
If my math is correct 24.2 Watts is 96.8 percent of Maximum. Can we truly call that conservative operation ??
...the 9 is a UL design so current is combination anode and g2. Anode dissipation is 25W and g2 is 8W for combined 33W. So the 9 runs the EL34s ~ 72% not 97%. Reasonably conservative if not the "ideal" golden ratio dontcha think?
I am not sure that that 8W is reasonable whilst the anode is run to 25W. The g2 grid posts are equipped with radiators at the top, which could allow this. U-L seems to leave EL34's run at high voltage to push their screens into having glowing spots, which is not what I'd call ideal.
While taking on design work in the past, U-L curves are not very much more linear than the pentode curves, though these curves have only been published at AB1 voltages. That would require a significantly higher load than is available to achieve Class A operation, and for that I will cap reasonably available at 10k a-a.
IMO, the primary benefit to U-L is a reduction in device output impedance; the plate resistance is lower...:) The next time I have the hoods up on my 813 monoblocks, I am going to run them full pentode, with their g2 at B+, with the plates i*R of the OPT primary less. Dropping the g2 voltage as current goes up *WITH THE 813* at 600V is not quite optimal. Or I will put them back to running the 4E27's...LOL
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.
...Problems can occur if all max ratings exist simultaneously: 800V/25W anode, 500V/8W g2 but that's far from the case here. Also note that these values are from the conservative "Design Centre" rating system, not today's more prevalent "Design Maximum" system . FWIW, data sheet calls out 30W combined anode+g2 in triode mode @ 500V. Overall, since subject amp is ~ 450V, 33W seems a reasonable max rating and 30W a conservative one.
Bottom line is that 24W in a 9 is a safe and conservative value.
Yep. I looked quickly. I forgot that it runs ultra non linear !!
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: