Amps Atma-Sphere M-60 Mono blocks OTL design

Driving easier speaker loads (high impedance) should be beneficial for all types of amplifiers and not limited to just tube power amplifiers.
Charles

+1  It continues to amaze me how many people don't get this basic fact!

Good!

How do you have them wired?

If you put the ZERO after the crossovers then the crossovers won't be operating the way they were supposed to. So before.

Now if I got this right, your speakers are home made? Could you build 16-ohm speakers? Then no need for the ZEROs at all.

It shouldn't as long as both are properly crossed over. The portions of the transformer's operation that are out of band with the crossover will see a fairly high impedance which will be transformed to a high impedance to the amplifier. Worth a try- you can't hurt the amp with an adverse load BTW. But I really doubt that it will yield much in the way of an improvement. OTOH, I don't know of anyone whose tried this, so report back :)

The Nirvana is a project to see how far we could push the technology. Its power transformers are spec'd differently, it has regulated DC filaments for the driver section and regulated high voltage for the driver section. The amp section is mounted on a subchassis and overall the chassis work is pretty over the top. Because of all the polishing, its been really hard to shoot photos of it!

We've got a lot of positive comments from customers using audiophile fuses.

The location on the amps that makes the most difference is the one by the power cord. That is also the one that is most likely to be blown, as a power tube failure can blow that fuse. The other fuses only blow if there is a serious service issue with the amp; over the last 30 years I can count those failures on one hand.

Because there are three fuses rather than just one, the sonic effect of the fuse is reduced somewhat but you can still hear it.

So far we've yet to hear about any hum problems with the Sophia or TJ Music tube. Al's report is the first I've seen.

We made a circuit change in our amps regarding how our filament circuit is handled about 12 years ago. That really reduced the hum issues caused by cathode/filament leakage; every now and then we run into a bad tube in the field causing a problem but its nothing like it was before we made that change. So I suspect much relies on the circuit design.

FWIW in our preamps the leakage would be a non-issue.

My forecast is the ZEROs aren't going to offer much in this installation.

Yes, but can the M-60 power these to rock music levels?

I play in a rock band. Audiophiles in general don't play **any** system to rock levels :)

^^ Yes! We have a number of customers that run ETs as well, and very successfully.

Doesn't using zero transformers defeat the whole point of an OTL amp? It's putting  a transformer in and makes it similar to a non OTL push pull tube amp.

'Defeat' as used in the question above suggests that an OTL is trying to do something that is somehow in opposition to a normal amp with a transformer?

It isn't. What we are trying to do is get as close to the music as we can.

The fact is that the ZERO has a very low turns ratio. Because of this it has very wide bandwidth- wider than almost any amplifier made, tube or solid state. You can't do that with a regular tube amp with an output transformer. You need to start with an amp that already has a fairly low output impedance, and a circuit that is not putting DC through the transformer winding.

The ZERO is really a problem solver- for example, here in Minnesota a lot of people want to use the M-60 with Magnaplanars. Usually they have smaller Maggies, but the M-60 is usually challenged when presented with a 4-ohm load. The ZEROs fix that and you don't loose the transparency of the amp. My rule of thumb is that when using the ZEROs you don't encounter any serious sonic tradeoffs. If the amp sounds better without the ZEROs then you just don't need them.

I think its better if the amp doesn't need the ZEROs. That usually means the speaker load is 8 ohms or more as in the case of ozzy's speakers.

@charles1dad , I think you might misunderstand. People hang on to their equipment, sometimes for life **regardless of the equipment**. To say that it "never goes one way" as you did earlier must be seen by a reasonable person as incorrect. That's all.

I was simply illustrating from personal experience that this is so.

The most common theme I hear when people actually tell me why they sell our gear (the second most common theme being they bought speakers our amps won't drive) is that they are downsizing or simplifying their setup (although its unclear what the latter means; if simplifying the signal path, the only amps I know of that have a simpler signal path are the First Watt DIY amps which have a single output transistor as the sole aspect of the signal path). As tube amps go we only have one gain stage which is about as simple as it can get.

These things never move in only one direction.

When I first got started decades ago it took me a while to get used to the fact that people would sell our amps and look for something else.

Over the years though the one thing that has really shown itself over and over again is how successful the initial setup of the amplifier installation actually was. If that was right (meaning: the right speaker load, short speaker cables, a competent preamp and front end to drive it) we found that owners tended to hang on to their gear. I don't know if owning it for decades until death counts, but if it does then the above quote is false.

This is why we developed our update and warranty reactivation policy, so owners of older gear could keep it current.

^^ Rear right and rear left are the input tubes, in parallel.

They are part of a single gain stage called a differential cascode, common in solid state circuits but unusual in tubes- when we first started using this topology we were the first to do that in a tube amplifier.

The middle tube is the top of said cascode circuit.

Therefore the two bottom tubes in the cascode circuit are the most important in the amp for sonics, followed by the top tube in the cascode (rear middle position). 

The front tube has no gain and is the driver for the output tubes. It controls the Bias and DC Offset. It should be a -GTA or -GTB if using NOS types.

I believe a more compelling and "competitive"  comparison would be the Line Magnetic 805ia.  This is a very highly praised 805 tube SET with 45 watt output.  Of course with these types of comparisons, final choice has much to do with plain old taste. Both of these fine  amplifiers will have their passionate and loyal advocates.

The speakers used are going to play an enormous role in the results! In this case, the Line Magnetic is at a bit of a disadvantage due to power, not just because its only 45 watts. The problem is that if you really want to know what the two amps can really do, the LM should not be run past about 20% of full power for its best results (otherwise the higher ordered harmonics tend to show up in its output). That's about 10 watts. Ozzy's speakers seem to be about 90 db or so; that really isn't going to allow the LM to strut its stuff without a lot of distortion unless you keep the volume down.

We have a lot of feedback from our customers on this sort of comparison.

OK: you asked. A lot has to do with the speaker of course, but the feedback we have from customers is that if the two amps are on a speaker that is friendly to both that essentially it is as you surmise.

But we get that a lot and not just with comparison to the Primaluna. That such is so is not surprising on two fronts; the most obvious being that obviously the positive comments are coming from customers that set the amps up right. The other is the overall design features of the amp; M-60 has full power to 2 Hz and can play bass quite well, owing to only one stage of gain and a direct-coupled output. With so little to go wrong its obviously very fast and more transparent without brightness. Triode operation. Fully differential. And so on (no single feature, but the combination of all of them makes for what the amp is). manufacturer caveat: we make the M-60 amplifier.

Yes.

Not that I know of. The tube testers I've seen don't make that distinction. If in doubt treat the tube as a GT.

@ozzy , make sure that 6SN7 that sits in the front by itself is a -GTA or -GTB. That particular position has more voltage on it and the older GT style tubes will have a shorter life in that position. FWIW the Russian and Chinese tubes marked 'GT' seem to be OK in that position.

It sounds like ozzy has gotten the amps to work in his system.

I have some tips:

The most important tubes for the sonics are the ones at the input of the amp. The precise location varies depending on when the amps were built (the M-60 has been in production in various forms since 1991). NOS 6SN7s get rolled into the amps quite often; the input tubes can be GTs. This means you can use Ken-Rad (of the NOS tubes they get the best marks from our customers), the Sylvania metal-base 'chrome domes' or the RCA Red Base (and its variant, the CBS Zalytron). But our customers have indicated that the Sophia, Psvane and TJMusic tubes seem to have all the best properties of the NOS tubes but without the weaknesses!

Keep the speaker cables as short as possible and the speaker connections as tight as possible- but don't tighten them so hard that you strip the threads or you will really regret it! Tight connections do sound better.

Ozzy sounds like he's running the amps on platforms. To test the efficacy of the platform, place isolation devices (squishies) beneath the amp and audition. If better, the platform isn't doing its job. If worse, then apply some sort of point system (Tip-toes, Stillpoints, etc.) to get the most out of the platform.

Damping rings can be applied to the 6SN7s but not the power tubes.

Power cords make a difference and the difference is measurable and audible. We've had excellent luck with the Purist Audio power cables and there are many more.

Fuses can made a difference too, but the most important fuse is the one by the power cord, and that one is the speaker protection fuse. If a tube arcs and does its worst, that fuse is going to blow, so if you have a $30 fuse in there it can get old pretty fast.

Of course in that regard,  our experience is that the Russian 6H13s (the Russian version of the 6AS7G) hold up a lot better than the Chinese tubes. Set up on the right speaker we've seen them last many years.

Never apply contact enhancers to the input connections or the tube sockets other than a light spray contact cleaner. Radio Shack sold a spray contact cleaner that was quite acceptable and you can also use the red and black DeOxit in a spray can. Anything stronger than that is asking for trouble.

If in the driver tubes, possibly. If some of the power tubes are blown out in one channel that could do it too. Check the power tubes against what you see in this article on the Support section of our website:

http://www.atma-sphere.com/Resources/6AS7G_Visual_Inspection.php

Some say the copper foil V-Caps take about 150 hours, be we experience them sounding very good right out of the box. However, with all the updates done on this set of amps, they're going to need some break-in time; as they break in they will become more lucid.

Since you're trying to nurse those old power tubes you might want to get some 3 amp slow blow fuses off of ebay or amazon. We sell them also.

As usual, the layout looks meticulous.

I assume the front 6SN7 nearest the faceplate still has the most effect when tube rolling?

Yes.

It seems to sound a lot better too. In particular the phono has more gain and less noise. My workers take a lot of pride in their work and usually sign the equipment after its done.

I notice with concern, the MP-1 now uses only two 6SN7's per side instead of five. Hopefully we're not losing any of the good things valves bring to the table. Besides lower noise, what's your thinking Ralph? Cheaper to re-tube? Just kidding! Hey, what about tube rolling? Have all my NOS jewels suddenly become worthless?

Ha! Actually the MP-1 uses three 6SN7s per channel. In the Mk3.3 version, prior to its release we had spent two years working on a Constant Current Source (CCS) circuit that could outperform the one in the earlier preamps (and amps). In the amps the old CCS drifted about 17 parts per million over an AC input voltage range from 100V to 126V. This in turn assured that the voltage amplifier performance stayed the same over the same range. In the preamp the same circuit was used but the preamp has severely regulated supplies so it was a bit more immune. At any rate, the new CCS is performing at about an order of magnitude greater performance so we were able to improve the differential effect in the voltage amplifiers of the amps and preamps; in the case of the MP-1 this meant that two 6SN7s got eliminated in each channel as they were CCS circuits. But the actual signal path tubes are the same.

To remove the bulb on one of our amps: unscrew the lens (lenses of other colors and texture are available). The bulb is an old-school bayonet mount and so you push on it and twist it 90 degrees counterclockwise and it pops out of the holder. To reinstall is the reverse process. The stock bulb is a type 47 which is a 6V lamp (we run it at 4.5V for greatly extended life and to dim it a bit); a 12V lamp can be installed.

The manual is available at this link:

http://www.atma-sphere.com/Manuals/

Just click on the one you want and its a pdf download.

Where's a good source for Russian and Chinese 6AS7?  

We stock both and both are available on ebay.

Usually when we ship a set of amps we precondition the tubes first and then play them. In this case since the amp was used it came with its own tubes. So we tested them and played them. Its not quite the same...

At any rate, over the years we've been researching how the tubes fail. Most of them fail immediately after shipment or else when they reach the end of their service life (which can often be 10,000 hours or more). The 'after shipment' part is what concerns us most.

What we've been finding is that most of the tubes fail due to arc-over and subsequent damage to the internal elements. It appears that the mechanism is the cathode coating, which is very much like a paint applied to the cathode (but in this case improves the cathode's ability to emit electrons; this is true of all tubes). The cathode coating can flake apart just like old house paint and if the tube is conducting the tube can arc as the flakes are conductive. Since many of these tiny flakes can be loosened during shipment, if the tube is subsequently subjected to a prolonged heating cycle most of the flakes that are occurring can fall harmlessly to the bottom of the tube and then the tube can provide normal service after that.

This applies to any in-directly heated power tube and is nothing unique to the tube we use. IOW running the amp in Standby for a few hours is a good practice anytime its been shipped, assuming the amp has a Standby function.

@ozzy ,

The Tung Sol name got bought out a few years ago and is now being used to mark Russian-made 6SN7s. Usually these are the same tubes as the Electro Harmonix but with a brown base. They are nothing like the old-school Tung Sols which were a nice tube.

We've had some very good comments on the Psvane, Sophia and TJ Music tubes; along the lines of 'combining the best of the NOS tubes without the weaknesses' which is a pretty strong statement. You do still have to pick them for low noise and microphonics though, but generally this is not a big deal in the amps. 

I do not think the ZEROs will be useful on this speaker.

The ZEROs should always be regarded as a problem solver (and for any amplifier). The problem is that the amp sounds worse without them, in a nutshell. If the amp sounds better without them then it should be a no-brainer that you don't need them.

M-60s do quite well on an 8 ohm load- the only issue I question is whether 60 watts is enough power, but since this system also employs a subwoofer and crossover, I don't expect that to be an issue.