Your journey with lower-watt tube amps -- Can a kit be good enough?

@hilde45 That link doesn't seem to work.

@hilde45 Excellent! If you change a component in your system and then you run into a buzz, this can be one of the causes. But in your case its not happening so no worries

@hilde45 Nice to hear you hare having success!

I prefer the JJ EL84s. You can get matched pairs at CE Distribution They have a Psvane tube that looks interesting too.

If the power cord is grounded, sometimes you can run into a buzz problem (ground loop) because the audio ground is also the chassis. To get around this you'll have to replace the RCA connectors (not a bad move since they are cheap anyway) and get a phenolic insulator to insulate the filter capacitor can from chassis ground. You that at this link at CE Distribution.

A 1/2 Watt 100 Ohm resistor is then placed between chassis ground and the audio ground.

This operation is a bit tricky since the chassis has to be modified (using a chassis punch) to mount the capacitor on its insulator. But I think you'll find it slightly lower noise.

Does your amp have the carbon composition or metal film resistors on the circuit board? The latter seem to sound better FWIW.

@hilde45 +1

Its amazing it did as well as it did initially! It's my surmise the prior owner never played the amp that much due to the error. So it might sound a bit better with some break-in. 

Highs pretty clear, though perhaps some slight roll off.

@hilde45 I was curious about this so I looked at the schematic of the amp.

The input resistor is 500KOhms (or 470K) to ground. With an input impedance that high, the amp can be sensitive to interconnect cables. If the resistor were changed to 100K, the amp’s character would be unchanged and it would still work with any preamp but the cable sensitivity would be vastly reduced.

So you might consider getting that resistor changed in each channel and see if you still think there’s a slight roll off.

BTW, the older amps have an input coupling capacitor which can usually be replaced with a bit of wire as the input circuit is at zero Volts.

@hilde45 I read that a few years ago and bought one of EFB kits for one of my employees who had an SCA35.

If you can shoehorn it in I think it would be a good move.

You can use a pair of such kits and run separate cathode circuits as well.

Or a bias supply could be built in. That might be more work.

At any rate the plate to plate impedance of the output transformer is fine for the EL84s despite what is written in that article, which is otherwise pretty accurate. According to the tube’s spec sheet, 17 Watts/channel is right for that tube.

Some of the newer kit versions of the ST35 use a choke in the power supply where the older original versions use a 50Ohm resistor. This causes the B+ to sag less (IOW better regulation), which reduces some of the issues brought up in that article. There is room for that choke in the original ST35 chassis. If you google photos of it you’ll see what I mean.

People like to see how far they can push it with this amp. Take a look at this:

Customized ST35

@hilde45 If you install that update I mentioned about a separate cathode resistor and cathode bypass capacitor for each channel, the cathode resistor value is 190 Ohms. I recommend a 5 Watt part so it will run cool. The bypass cap is then 50uf. They can mount inside the chassis no worries.

@hilde45 I'd go with a set of JJ EL84s before Sovteks or for that matter any Russian made version.

Have the tubes been tested? If any are bad all bets are off.

If you have level controls on your speakers it might be a good idea to run pink noise through the system and using a spectrum analyzer app on your phone, adjust the controls slightly for best results. This amp has a lower output impedance than any tube amp (such as an SET) that runs no feedback so if your controls (if you have them) were optimized for an SET they won't be right for this amp.

Your amp in the photo looks fairly new. Apparently there's a resistor option on this amp; do you know which kind your amp uses? It looks like they might be the carbon film resistors. IME the metal films sound better.

If the amp has been sitting unused for a while or is brand new, it may need some break-in time.

I doubt its actually rolled off- this amp has full power bandwidth well past 50KHz.

@bruce19 There's a kind of ceramic cap called an 'NP0' (for Negative Positive Zero) which can sound surprisingly good. They are available with leads and in the value you need, are about the size of a small bead. By ceramic cap standards they are expensive, costing a few dollars each. You might look into them; we've compared them to a number of high end coupling caps and heard no difference. I don't think they are available in values above about 0.1uf. But they are available in Voltages as high as 630V. You only need 50v though so the part will be tiny. Here's a link from DigiKey:

https://www.digikey.com/en/products/detail/murata-electronics/RDE5C1H273J2M1H03A/10703951

@bruce19  You could install a coupling capacitor between the source (perhaps a preamp) and the SET, with a value calculated to have a -3dB point at the desired crossover frequency.

Frequency in Hz = 1,000,000/Capacitance in uF times Resistance on Ohm time 2Pi  (Usually when this formula is used the one million is a 1, but the capacitance is in Farads, which is too large to be easy to use. So I do the formula this way so I get practical audio values)

So if you wanted 60Hz as the crossover frequency and the input impedance of your amp is 100KOhms, then the capacitor would be about 0.027uf (that yields a 59Hz -3dB point).

This cap can be tiny since it does not need to handle much Voltage. So you might be able to install it inside an RCA connector of the interconnect cable.

@mclinnguy Thanks!

So what is actually simpler? Many say SETs are. But they have problems that are very difficult to solve like the elliptical load line I've mentioned several times. I think they are just too simple.

In a conventional push pull tube amp the problem of core saturation in the output transformer is solved (so no elliptical load lines). This allows the the transformer to have much higher inductance at low frequencies and generally lower distortion at any frequency.

If the PP output circuit is biased class A1, it induces no more noise in the power supply than an SET. If the circuit is fully differential then the zero feedback distortion signature can be just as pleasing if not more so then that of an SET.

What I'm getting at is there is a thing I like to call 'elegance'; a simple solution to a difficult problem.

The original Ultra Linear concept was elegant. It was so popular that many sought to get around the patent by simply moving the UL tap to a slightly different spot, sacrificing some linearity to avoid paying a patent royalty. This was so prevalent that most UL transformers made today still do not conform to the UL patent, despite it being long expired. That's why some people like UL designs and others don't- they are not all equal. But the original idea worked quite well and displayed that quality of elegance. 

In case those reading these words aren't familiar with the concept, UL operation allows a pentode power tube, either single-ended or push-pull, to have the linearity of a triode combined with 90% of the power of a pentode while being much easier to drive than a power triode. Anyone doubting this probably should read the original patent issued to Acrosound in the 1950s (David Hafler, one of the inventors, left Acrosound for Dynaco; knew how to break the patent and did; the UL patent was never licensed to Dynaco). 

One thing which seems to hang on this answer is when and whether the general audiophile precept that "simpler is better" is true.

@hilde45  I think it was Einstein who pointed out that a thing can be too simple to do its job well.  I prefer simplicity myself and like to keep things simple if I can. But while a thing might appear simple, from just viewing it you may not see how much is really going on.

For example to get a 300b to sound right it has to be biased properly. That requires one to generate a load line so that you can see if the output transformer used is going to load it correctly, in addition to the operating Voltages. You can't see that just by looking at it.

A self oscillating class D amp can be a fairly simple circuit too. Looking at one, you can't see that the math behind it is pretty complex.

Of course, a lot hangs on what is meant by 'simple'...

explain in a bit more detailed manner [my own ignorance] how "feedback" sort of identifies, isolates, and/or removes these other things that are "not" the signal.

@decooney The various influences described above (resistors, caps, transformers, tubes, et. al) are part of why the amp makes distortion.

Feedback works by taking a small amount of the output Voltage and applying it to an earlier part of the circuit where it is out of phase with the audio signal. When the two are mixed, a corrected signal is generated which has the corrections needed to make the amp put out an undistorted signal.

In theory.

In practice feedback can generate distortion of its own. This is especially true if the feedback is mixed with the audio in a non-linear way, such as inside a vacuum tube or transistor. This is very common; in addition to harmonics the non-linearity of the feedback node can generate intermodulations as well. 

Your tech friend is right. Too many designers just use a resistor and call it good. If a lot of feedback is used, this can lead to instability. The amount of feedback used and where its applied makes a difference too. If you want to do it right, this complicates the successful amplifier design in a lot of ways since the math for a second or 3rd order feedback loop isn't trivial.

 Power supplies of all vintage tube amplifiers from 50-60x look so miserable with such small capacitors.  It is especially true for stereo amplifiers when PS is used for both channels in parallel without any separation. I can say for sure such a PS will not work for a SET amplifier without feedback even if it is just 2 watt power.   

How does it work in all these vintage amplifiers? Does feedback make them less sensitive to PS or the reason is the push-pull topology?

@alexberger  Feedback improves power supply rejection and crosstalk; ideally anything that isn't the signal. For example, in an LP mastering situation, the feedback winding on the cutter head is essential to allow channel separation. The stereo LP would not exist without it. 

Such a power supply as you described above will work fine in any SET that is running class A1 on the power tube, since at any signal level current draw is constant; IOW from idle right to full power. Consequently this statement:

The issue with small capacitors is not the 120Hz noise from AC , but voltage stability during playing of load and complex music.

-is false.

Boosted capacitance does help with a stereo amp or even a mono amp if the amp is push pull and operating class A2 or any form of class AB (IOW where current draw varies with power), so as to reduce intermodulation.

In all cases, the proof of the pudding is to put an oscilloscope probe on the power supply rail and run the amp up to full power at a variety of frequencies to see what sort of noise is present in the power supply. Since the output circuit isn't linear, power supply noise can intermodulate with the audio signal in the output section.

@hilde45 The article you quoted was true back in the 1980s but is not true now. The real issue isn't that we can't measure it, its knowing what the measurements are telling us. That knowledge is in short supply! I've been explaining exactly why SETs sound the way they do. Again, in a nutshell: They shine when not presented with bass or asked to make much power.  I've already explained why they sound good when they do so.

Harley's mistake was simply not asking amplifier designers why SETs sound the way they do. But if he had done so he might not have had an article to write I was online pointing this out when this article originally published a few years back.

I posit at least part of this quandary is a function of low parts count in SET. SET doesn't have to negotiate multiples of caps, resistors, transistors, what have you, and if point to point wired no circuit board. I can hear sound quality differences with a changing out a single resistor, cap or hook up wire, this goes to show how transparent these amps are. I used to have to change out multiples of these items in other amps to exact a change in sound quality. This suggests a signal passing through a less circuitous route is less changed/affected/contaminated, in other words more pure to the source.

@sns This statement comes to a false conclusion. The reason you hear differences due to components is not because SETs are particularly transparent (due to their having the highest distortion of any kind of amp made, and because distortion tends to obscure detail, its arguable they are the least detailed amps made), its because they lack feedback.

Feedback helps the amp reject that which is not the signal. Artifacts from wire, resistors, capacitors and transformers are not the signal, as well as noise from poor grounding, layout problems, line Voltage and so on. So when there is no feedback, everything (and I mean everything) makes a difference.

The problem of course is whether or not the feedback is applied properly and we've all heard amps where it wasn't. I've no doubt that many might prefer an SET over such amps! But its usually not due to detail- its (IME, IMO) due to amps with poorly applied feedback not sounding particularly musical on account of harshness or the like.

Yes this Marantz 8B sounded more clear then McIntosh MC30. But despite Marantz bass wasn’t bad, MC30 bass was on the other league. It was also a big difference in complex music reproduction. 8B was good on very simple music but the sound became muddy and congested with bigger number of instruments and voices. On the other hand, the complex music wasn’t an issue for MC30.

@alexberger Sounds to me as if the modifications didn't have the desired effect. Feedback is a bit more complicated than it might look on a schematic. The Marantz 8B has a more sophisticated feedback design than most amps you see from that era; IMO/IME its unlikely that a technician 'adjusting' it would be successful in getting the amp to actually perform better.

@hilde45 Thanks!

Didn't have any particular issue with the amp, just blah, lack of transparency, resolution. This compared to two 845 SET amps I was running in that time frame.

@sns  The Merlins are not a good match for any SET unless you are using them in a small room. They are 88 or 89dB 1 Watt/1 meter and 6 Ohms (so the 4 Ohm tap should have been used). You need a good 50-60 Watts for them in my experience.

The VTA ST70 is something different from the ST70; it didn't measure as well as the original FWIW. 

@decooney , @hilde45 , One weakness of that amp is seen in the schematic.

The 4 power tubes share a single resistor (bypassed by a capacitor) to set up the bias of the output tubes. There really should be a resistor and cap for each output tube pair. This was a space and cost saving move on Dynaco's part.

That's a pretty well known issue in this amp and the SCA35 which uses the same transformers. By separating the channels in this manner the amp performs and sounds better. 

@hilde45 Nice! I'll be interested to hear your comments. I saw that amp on ebay as well and really had to think about buying it just so I could see how far I could push it.

@sns To be clear, the Klipschorn rolls off sharply below 45 Hz.

The older Klipschorns are not designed to be driven by an amplifier that has a low output impedance like an ST70. You'll know if your speakers are like this (the Klipschorn has been in production since the 1950s which is a bit of a record) if there are level controls on the back for the midrange horn and tweeter.

SETs are a technology which I call the Power Paradigm. PP amps like the ST70 are Voltage Paradigm technology since the ST70 employs enough feedback to allow it to behave as a Voltage source. IOW this has to do with the output impedance of the amplifier; whether or not its low enough for the amplifier to behave as a Voltage source. If your Klipschorns have level controls, essentially they are Power Paradigm technology also and there will be tonal aberrations if you try to use them with an amp that is Voltage Paradigm.

For more on this topic see http://www.atma-sphere.com/en/resources-paradigms-in-amplifier-design.html

So, a good SET with matching high sensitive speakers will be a huge step over any vintage push-pull amplifier including Dynaco.

@alexberger I suspect your sample size is a bit small to be making that conclusion. For example I think the ST35 is more musical than the ST70. I also think a properly refurbished Citation 2 is one of the better amps from the classic vintage era. I’ve always felt the Mac stuff from back then (my exposure limited to refurbished amps, M30, M240 and M275) to be less articulate but that’s me and entirely anecdotal.

Fisher made some very interesting monoblocks that featured a damping control. A pair of those allow you to compare on older speakers like the Altecs by dialing the damping to be a bit lower, since a Mac or even an ST70 will have an output impedance that is too low to sound right on the classic Altecs. EV had a similar feature on some of their amps too.

@retiredaudioguy You might want to keep an eye on any coupling caps you have that are oil filled. I’ve found over time (we’ve tested a variety of them) that they can often develop a slight amount of electrical leakage across them (IOW not entirely blocking the DC Voltage they are supposed to block, sort of like having a resistance in parallel with the cap) which can throw off the operating point of a power tube! The exception to this is the ODAM parts from VHAudio.

Do you think the issue is not voltage swing of driver but the current and power it should drive on capacitive load.

@alexberger Generally speaking, yes. Specifically you want to be able to swing the Voltage without distortion. To do that the source that does the swinging must be able to provide the current needed, if any. Power triodes tend to have a bit of grid current anyway even if you aren't driving them positive with respect to the cathode; some of this is due to grid capacitance.

The bottom line is a gutsy driver is needed when driving power triodes of any kind. That is why RC coupling is not the best solution- you use either an interstage transformer or direct-couple; the latter being lower distortion since its a much easier load to drive for the Voltage amplifier (cathode followers tend to have a very high input impedance).

Transformers make distortion of their own even if properly driven and loaded and they limit bandwidth. So if you want the best transparency you don't put them in the circuit. 

@larryi Decades ago I put myself through engineering school by working as a repair technician at Allied Radio Shack. I saw equipment that clearly was not built with the ability to easily repair it in mind. Electronic components fail no matter how well you vet them or how well you rate them in the circuit. IMO/IME easy repair should always be high on the list when laying out a circuit. 

If you do it right and use good quality boards the repair time is about the same (if you do a neat job) whether circuit board or hand-wired.

@alexberger 

The problem of 300B they need a driver with 150 volt peak to peak voltage swing. For 845 you need much more voltage swing.  

If you direct-couple with a CF that's no problem at all! Here's something to consider:

The grid of triode power tubes is the most capacitive of any kind of power tube. Triode power tubes are also very low mu (gain). For example the 300b is a bit over 2. The 6AS7Gs we drive in our OTLs is only 2.

So driving such a load from the plate of a driver tube is a task- hence the common use of an interstage transformer. But that part is expensive if its done right and it makes distortion and limits bandwidth.

When you use a cathode follower its input impedance is really high so its a lot easier to drive than a power tube and easier to drive than using an interstage transformer. So its a lot easier for the Voltage amplifier to drive it without distortion and the driver itself is inherently low distortion as well.

Since you are into DIY you might consider trying this. Its not that hard and like I said, the tube section to do it and the minus supply is usually less expensive than a good interstage transformer. As a general rule of thumb whatever the B+ Voltage is that you use now for the driver, if using a direct-coupled cathode follower you'll need a B- of equal absolute value. That will allow it to have the Voltage swing you need. A single 6SN7 section can easily swing 150V or more no worries. We've been doing it for decades. Usually you can obtain the negative supply if you use a bridge rectifier. That requires a power transformer with a center tapped secondary. 

845 is impossible to drive without IT! You get more driver tube distortions than 845 output stage distortions. You also can't get a good deep and fast bass.

@alexberger This statement is entirely false (and is what IT manufacturers would like you to believe).

If the driver is a cathode follower and direct-coupled to the grid of the power tube you'll have no problem with drive and less distortion than you can get with an interstage transformer. SET designers don't do it because you have to have a B minus supply for the driver tube and they don't want to spend the money although a good interstage driver transformer will cost more. The other reason SETs designers don't do it is they didn't think of it!

You'll recall I mentioned this technique on another thread on which you were active on this website.

For the record we've been driving power triodes in exactly this manner for decades (47 years to be more exact). The driver has an iron grip on the grid of the power tube and can easily drive it into grid current with good linearity. It also prevents blocking distortion which is a problem for any tube amp which employs coupling capacitors between the driver and grid of the power tube so you get instantaneous overload recovery.

You might try it since you're into DIY. If you have any questions about how to go about it, email me.

Just my two cents, but having owned multiple ST70s and ST35s, I like the ST35 much better.

@hilde45 This is a comment worth exploring! The ST35 uses a triode Voltage amplifier while the ST70 uses a pentode. In addition, the smaller output transformers of the ST35 have wider bandwidth.

The input of the ST35 uses a 7247 tube, which is a 12AX7 section and a 12AU7 section in one envelope and is in current production. The 12AX7 section is the Voltage amplifier and the 12AU7 the driver. Since the 12AX7 is far more linear than a pentode (and has the most gain of any miniature triode) when feedback is applied, the tube distorts the feedback less (than a pentode) so the feedback is more able to do its job properly. This results in less higher ordered harmonic generation, so this is a much more interesting and relaxed amplifier to listen to and will fulfill your lower powered amplifier interest.

If one of these is properly built/refurbished it will give any SET a run for the money! It will play bass better, be more revealing (less distortion) and more neutral (less distortion).

And there’s a kit! Or there was. Their website says out of stock.

Go with the metal film option. Carbon resistors are really terrible.

Atmasphere your viewpoint in this discussion repetitively shuts down people learning and exploring. The whole discussion about bass, about distortion, and many of the other salient points you’ve raised, while one hand accurate, are on the other hand wholly irrelevant.

@yaluaka  The way you write suggests to me that we have very similar goals. When I put on an LP I don't want to be thinking about the system. I want it out of the way so I get the full experience of the music and nothing else.

I'm really trying to save people from flushing perfectly good dollars down the loo rather than trying to shut down discussion. You'll note I suggested some simple solutions for getting the most out of any SET on this thread. I can outline them again if you like.

How relevant my comments are depends on what you play for music. I have recordings I recorded, so having been there when that happened, I know how they are supposed to sound. I enjoy electronic music and some of that has really deep bass. So its only irrelevant if you limit the kinds of music to which you listen. If you only play chamber music or light jazz or folk you'll probably be fine. But if you have an original LP pressing of Hendrix's Electric Ladyland, King Crimson's In the Wake of Posiden, ELP's first LP and so on you really won't get to experience what those records are all about with an SET because of the bass issue (if nothing else). If you can find a copy of RCA's Soria series recording of Verdi's Requiem its the same thing. They really didn't compress that LP and it can bring any SET system to its knees in very short order (side one, track 2 if interested).

BTW its not just SETs that have problems making bass properly. A lot of early solid state stuff does too. That issue with solid state went into the early 1970s until silicon transistors got good enough that complementary symmetry was finally possible. A lot of those amps are nowhere near as musical as a lot of SETs not as if that's any revelation...

Maybe there’s not as much bass etc, but for me the way to listen to music, is to listen to music not equipment. Listen, don’t analyze. I find tube triodes to give me immense musical pleasure despite their drawbacks.

FWIW I've been building triode power amplifiers for over 47 years. Its not triodes that have the drawbacks I described; triodes can go right down to DC if you want. Its the output transformers in SETs that have the limitations. David Berning is the only one I know of whoever figured a way around that and unsurprisingly his amps get very good comments. 

How many maximal dB do we need for classical, rock, jazz music?

It looks like the distortions from 50Watt heat on the speaker's voice coil and crossover parts should be much more horrible then 10% SET distortion on maximal power distortions. 

@alexberger The point isn't to use all that power! In the best installations, the amplifier should be loafing. No matter what kind of amp you have, the harder you make it work for a living the more distortion it will make. IOW the point is to make clean power at normal listening levels. When its clean, it won't sound 'loud'.

I don’t agree with them, for one very important reason, this is a hobby, it’s so much fun to explore, it’s not have people tell you you won’t like something when in fact, if your like me or many others, you actually will.

@yaluaka The stuff I said about SETs having troubles with bass (there are other problems too but I'm focusing on just this one for this post) is very real. I certainly agree that the exploration can be quite fun (I've built a number of SETs since the early 1990s and did it solely for that reason). But the engineering issue about bass has been known for many decades (google 'elliptical load line' and you'll see what I'm talking about). That is why I said that the most successful SET installations (and some of the most avid SET lovers I know) have some means to keep bass out of the SET. When you do that the mids and highs get a lot better! Its not subtle.

Here's the situation: I have new speakers, 97 db. Trying them with lower watt tube amps (45/211, 300b, etc) seems generally wise. I am attempting to borrow some from audiophiles in the area. 

@hilde45 Unless your room is a smaller one, SETs using the tubes you mention here won't have enough power if you want realistic volume levels. They might sound 'loud' when playing such amps but if you use a sound level pressure meter (available as an app for most phones) you'll see they really aren't playing that loud- their 'loud' quality is brought on by distortion rather than volume, and how the distortion interacts with the human ear.

My speakers at home are 98dB and I found I needed more like 50 Watts; less than that and the amp seemed to strain a bit unless playing at lower volume levels. I don't play stuff all that loud FWIW, but if I go over 95dB I want the system to remain relaxed and effortless.

It won't matter if the amp is handwired or circuit board as far as sound is concerned, although a handwired amp can be updated a bit more easily and might hold up better in a tropical environment (such as on the coast in Florida).

I think you can still get Dynaco kits. A properly refurbished and/or properly built ST-70 can give any SET a run for the money, especially if given the benefit of good quality parts. For that matter we make a little 5 Watt stereo amp that can run circles around any SET that is near its output power. The 'magic' of SETs is mostly a made up story, based on the simple fact that when the SET is compared to a PP amp, the playing field is anything but level. When the PP amp has the same power rating and is built properly, the advantages of SETs go away.

SETs have a lot of weaknesses. I can go into them if you are interested; not going to list them all here. But two big ones are: you have to be very careful of the loudspeakers used as SETs usually have no feedback so the normal Voltage rules the typically govern speaker design don't work for them; in addition SETs only make about 20-25% usable power so the speakers have to be more efficient than most people think. So about 99% of speakers made are inappropriate for them.

The other problem is bass. SETs aren't good at that because the output transformer typically lacks the inductance needed to make bass, on account of having something called a 'cut core' which is a saw cut in the core of the transformer, put there to reduce distortion caused by DC flowing thru the transformer to feed the power tube. The lack of inductance means at low frequencies the tube is driving something very nearly a short (as far as the tube is concerned; put another way its driving the DC resistance of the winding of the transformer which is a couple orders magnitude lower than the impedance the tube is meant to see) and that's hard on the tube.

So for best results with SETs, its a good idea to prevent bass getting into the amp. The most successful SET installations I've seen and heard all used a different amp for the bass and had some sort of crossover to prevent the SET from having to deal with the bass.

The one exception to this problem are the Berning SETs, which use a patented system that gets rid of the traditional output transformer, using instead a radio frequency method of coupling between the power tube and the loudspeaker.