What's the best EL84 amp?

@dhcod They might start popping-up again sooner than later. Keep an eye out for them. I wish there was some way to add an image to give you more clarity, but for now you'll have to trust me on this.

I just saw that WTB ad. Will be in touch.

@dhcod Roger made some improvements in the MkII version and laid the groundwork for more improvements in the amps circuit design prior to his death. Note that there are multiple MkII versions (Roger felt as the designer he could make changes anytime he wanted), where the MkI retained the same circuit and specs throughout its run. You will find those who feel the MkI is a better amp, just as there are those who felt the RM-9 MkI was better than it's MkII version.

I tell this to anyone who is interested in buying a used Music Reference amplifier these days. Inspect the interior very carefully. If necessary get pictures of the interior and send them to us to verify the amp has not been modified beyond reason. Lately we have seen too many Music Reference amps come into the shop for repair that were modified to the point where putting them back to original spec wasn't worth the cost to the owner. In fact we now refuse to service amps modified beyond reason.

@bdp24 More people than one would have expected have told us they felt the RM-9 MkII sounded more like a solid state amp than a tube amp. It was certainly a lower distortion amp, with an upgraded power supply featuring added filter capacitance and chokes, but 8 x EL-34s for output tubes is 8 x EL-34s so go figure.

...with huge transformers. I wonder if that's the difference?

@dhcod Yes, sort of. After dealing with the large potted transformers on the RM-9 which accounted for quite a bit of it's weight, Roger wanted a tube amplifier with a small footprint that he could easily carry around. So for the RM-10 he made much smaller transformers and compensated for them with networks in the circuit. The amplifier is extremely flat from 30 Hz to 20k Hz, but things start easing up below 30 Hz, so that might be what you are hearing.

At the time of it's design 35 watts from a pair of EL-84s per channel was double what other designers at that time were getting out of two pairs of the tube. Putting 700 volts on the plates was unheard of, until Roger did it. This is why the amp punches well above it's weight and has a pretty nice set of balls.

As someone who worked with Roger Modjeski for the last 5 years of his life, and to who (along with a couple other long time friends) he entrusted his business and legacy with upon his passing, I see I need to clear the air here.

A lot of facts about what Roger did and how he did it have gotten twisted over the years, to the point of becoming myths. Roger tried to set the record straight a number of times but finally gave up. I think he eventually came to view the myths as a source of humor, to the point of proliferating them. Roger and I spent his last few months speaking daily at length about many things audio (including the myths vs. reality) and personal. We listened diligently to his interviews with audio luminaries like Julius Futterman and Saul Marantz. I have all his notebooks and designs (at least the ones he documented) that I have been reading and studying so I think I have a pretty good idea of the man and his legacy when it comes to audio.

@bdp24

I got myself an RM-10 for a specific application, one for which the amp is perfect: the Quad ESL. Roger used that loudspeaker and the 16 ohm LS3/5a as his speaker load in the design phase of creating the RM-10, the Quad a notoriously difficult speaker to drive.

The Quad ESL was never used in the design process or as the load for the RM-10 and it’s debatable that his Chartwell LS 3/5a speakers were either. The Quad just happened to be the speaker Roger was using in his system when the amp was designed so after playing it on them just to get a listen to it he felt it was a good amp for the speakers, but a bit lean in the bass.

I have always used subs with my Quads, as did Roger.

Maybe you did but not Roger, although he certainly felt that it would be a benefit to do so, which is what he told me as he sat next to me listening to my RM-10 power my Quads. So next day at his shop we made up some woofers, I grabbed a solid state amp off the shelf, and Roger gave me a Beveridge RM-3 active crossover so I could biamp the Quads in my system and I never looked back.

In the 1980’s (I believe it was) Roger was hired by Harold Beveridge to design and build the tube amps included in the Beveridge ESL.

It was the late 70’s when Roger worked at HBI and this was probably the classic of all the myths. Roger never designed or built those direct drive amps. He did test all of them which is why you will see his name inside them, but design and build - no. Roger did design the Beveridge RM-1 preamp, it’s RM-2 power supply, and the RM-3 active crossover. Roger remembered this period of his career fondly as he got to focus on what he truly loved which was design work. As he often told me, at HBI I designed it, someone else built it, and someone else sold it.

Roger’s final product was not just his own ESL loudspeaker, but one with no input transformers (the cause of a lot of non-linearity).

Not by a long shot.

@bdp24 Sorry if you felt I was churlish. Just trying to set the record straight and it's probably something I picked up from Roger himself after many years of listening to him trying to set people straight about all things audio. He certainly didn't suffer fools gladly.

The ESL direct drive amps designed by Roger were never meant to be a product per se. The first ones were actually designed as a custom project for some Acoustat owners down under who wanted to replace the pitiful step up transformer that was originally designed for the various Acoustat full range ESL models. Roger appreciated Strickland's ESLs and they influenced his ESL designs. Roger was also very familiar with the Acoustat direct drive amp and designed a tube input stage modification for it that a few folks have benefited from. So he designed an all tube direct drive putting 5000V on the panels without that pesky interface for those very lucky Acoustat owners.

When Roger decided to build an ESL of his own, the earliest of which I knew of was around 2003, it wasn't used with a direct drive amp and it wasn't full range, it only went down to 100 Hz (as it was with all his other ESLs) and was biamped. I remember going over to his house to pick up a box for my RM-9 and he invited me in to listen to them with none other than Kavi Alexander who had dropped off his Beveridge ESLs for Roger to fix. It wasn't until he built the direct drive amp for the Acoustat guys that Roger built one for his ESLs (which required less voltage, I think around 3500V).

I would have to say Roger's last "product" was the RM-9 Special Edition. Limited to 20 pieces it is an all point-to-point wired circuit using an additional pair of driver tubes, 6BQ7s at that, and adding another 25 watts of power. Just about everything else was the result of custom orders and small batch efforts (ex. EM7 amp). Even the OTL 1 amp had only 6 pieces produced (on a RM-200 top plate where Roger didn't even bother to hide the RM-200 logo).

Roger's idea of "product" was very different than industry norm. He felt that as the designer he could make changes whenever he wanted without having to document or re-badge the designs. This is why there are 4 versions of the RM-10 MkII sitting out there. He kept most everything in his head but in the final year we pulled enough out of him to hopefully bring forth some posthumous designs that audio enthusiasts can enjoy. So yeah, in more ways than one that ESL direct drive amp was not Roger's final product by a long shot.

@dhcod Well you would have had to jump on the 405 so doubt we passed each other. At the time I lived in Santa Barbara 3 miles from Roger :) I have the only remaining pair of ESLs with the direct drive amps. One of these days I’m going to have to hook them up. Roger figured out a way to increase the dispersion of the treble in the panel to create better imaging. Roger mentioned to me it was something Jim Strickland was onto as well but Acoustat went by the wayside before the work could be finished.

@niodari The Russian military EL-84s (Soviet Era) and Yugo EL-84s were Roger’s preferred tubes for the RM-10. Both are excellent choices. I’m very curious about the Manley EL-84 amps and think a set of Mahis may be in my future.

@pinthrift We're around to fix the Music Reference amps but will no longer touch them if they have been modified or have had unauthorized parts used in them. The latter would include after market "boutique" fuses.

For the record Roger updated the fuse specs to 250 mA ceramic (not glass) fuses from the original 160 mA spec. We have them here or you can get them from reputable sources online such as Mouser or DigiKey. Roger's preferred fuses were made by the Littelfuse company.