Jump to content

Recent trashing of the VTA ST-70


Idontknow

Recommended Posts

On 4/16/2022 at 4:28 AM, Schu said:

Tests great... sounds terrible.

Or, in this case

Tests terrible... sounds great.

Tests terrible...sounds great....OK whatever floats your boat.

Tests great....sounds terrible....time to get your ears checked:)

Link to comment
Share on other sites

On 4/16/2022 at 7:28 AM, Schu said:

Tests great... sounds terrible.

Or, in this case

Tests terrible... sounds great.

There is truth in what Schu has quoted above, but it is all relative. There is a long history of equipment that does not have the best bench test results, but the sound quality is very listenable and in some cases, there is a cult status of said equipment. 

 

In other cases, the equipment bench tests are great, but the amp is cold, sterile, analytical, etc and in general, people do not prefer the sound. 

 

Link to comment
Share on other sites

2 hours ago, Curious_George said:

Here is a link with a little background on the VTA ST-70;

https://www.audiophilenirvana.com/audiophile-equipment-reviews/dynaco-st-70-timeless-classic/

 

 

 

Thanks I was reading about what changes have been made.

 

If they really wanted to make the front end better they could have done a much better job in my opinion. They added a tube and got worse performance, that's because split load has great balance where the LTP can but needs either constant adjustment or a current source load at the tail, it doesn't look like they are using an active load for the tail of the PI so that was a poor decision. If they wanted to keep the same all triode arrangement why not use one 12AX7, one triode section per channel for voltage amp, cascode a couple depletion mode Mosfets as an active plate load, that combined with a bootstrapped LTP 12AU7 which will have a very high input impedance the voltage amp will reach mu for gain and  have extremely low distortion. This will give you a ton of open loop gain to bring the feedback back up to where it was with the original. You could even ditch the active plate load for the 12AX7 and still have enough gain to increase feedback substantially.

 

I personally feel the pentode voltage amp is part of the ST70 vibe. Switch to a 6BL8 or similar and run the front end the same except insert two 12AU7 driver stages (common cathode) ala Williamson circuit. Or just ditch the 6BL8 and use 4 12AU7's in the front end for the actual Williamson circuit which IMHO is one of the best push pull circuits around and it's really simple.

 

It's possible I have heard a VTA ST70 but I can't recall, I certainly haven't had one in my home and critically listened vs the original to see how they compare. I just find it funny how they made an "improved" board with worse performance. For what they have at their disposal 1% THD at 10 watts isn't hallmark of a great design with a push pull EL34 amplifier..

 

I'd also go ahead and ditch the ultralinear to save on power tubes, run the screens around 350v and add a zobel at the output to keep the load flatter on the pentode at higher frequencies.

 

The magic of the ST70 was that it used just one preamp tube and got good power and performance. It just seems odd to make it more complex without achieving better performance.

 

The last EL34 push pull amp I made had less than 1%THD at 20 watts WITHOUT FEEDBACK, and with feedback I got .05%THD  at 20 watts

Link to comment
Share on other sites

7 minutes ago, John Warren said:

The PC board layout was done by newbie.  

 

 

I concur, poor layout for more than one reason.

 

Pin 8 looks tied to a transistor, which I am sure is also tied to pin 3. Is that a TO-92 package? I am curious what the tail circuit looks like with just one transistor, I typically need at least two get a high enough AC impedance to be worth it. A TO-220 package would indicate a 10M45S type current source package which will give decent results but it doesn't look like that package to me.

Link to comment
Share on other sites

Little searching found it's a LM334

 

Have a look at the datasheet output impedance vs frequency, it drops like a rock after 10kHz and won't have great high frequency performance. Not a very great selection, at 20kHz it's not going to balance the output like it would at 1kHz, curious what distortion is like vs frequency.

 

Some simple cascoded depletion mode FET's will achieve a source impedance in the 1000Meg range, a little better than 1Meg at 10kHz for the LM334, but I'm sure they just used the simple adjustable CCS because well it's easy to apply, small and cheap and that's it. It's basic electronics knowledge, a perfect constant current source has a theoretical infinite impedance, so the higher the source impedance the better it will work. Regardless if they could have done that better it's still nice to see some modern circuit design going into the amplifier. If you think about it, tubes are high impedance devices, so it's natural thinking to improve upon the older circuits we can implement more modern high impedance loading techniques. I think it has to do with some myths that if there is a transistor ANYWHERE in the amp it's a hybrid, but all we are doing is using silicon to HELP the tubes perform better.

 

 

 

lm334.png

Link to comment
Share on other sites

@Idontknow Good video!  What he could have mentioned which may or may not have helped was the music theory and harmony aspect of distortion, I like bringing this up because it helps explain why some distortion is better/worse than others, he only mentions how far the frequency is away from the fundamental being the reason it's more audible but it's a little more complex than that, as I explained it has to do with harmony between two frequencies, how they interact with each other and how we interpret the two notes together as harmonious or dissonant. For example the 4th harmonic of 1kHz is still the same musical note two octaves up, this means the 4th harmonic (4kHz) although it's further away from 1kHz than the 3rd harmonic (3kHz) is more harmonious and less audible compared to the third harmonic. The 3rd harmonic is a musical 12th which you can think of as a musical 5th. In music chords are made up of triads, three notes of a scalar form, 1-3-5. One major aspect of rock music is the use of the "power chord" which omits the 3rd note in the triad and just leaves you with 1-5 notes. These two notes sound very powerful together, hence the "power chord", it has an aggressive powerful sound to it. So a push pull amp with 3rd harmonic dominant is going to sound much different than a single ended amp with the same amount of THD but the distortion is 2nd harmonic dominant with the SE amp, the latter being much more neutral and cleaner sounding.

 

I don't dismiss the SINAD test scores, I merely use it like any other tool so long as you understand it's limitations as this video points out. Personally for me a DAC is something I will look more at SINAD ratings vs a power amplifier.

 

Another thing he mentioned that I liked is the louder the volume the harder it is to discern these distortion differences, I'd like to add upon his remarks beyond how the ear works but looking more at the larger chain of the signal. The point I'd like to bring up is the distortion from your speakers, the louder you push your speakers they will contribute far more distortion than the amp will so the amps distortion at higher output powers will be far lower and less critical.

 

This is why it's hard to just give anyone an easy rule of thumb on how much distortion is too much, it's just a complex concept that needs to be thought of as such. So I honestly can say without much hesitation that .5%THD where say it's second harmonic dominant is less audible than .1% of higher order harmonics say 3rd and above. I may even go as far as to say 1% of second harmonic is less audible than .1% of higher order harmonics.

 

So it's a matter of just looking deeper into the SINAD score to gain more insight into how good or bad something is. Of course if the device has a very good SINAD score you can just immediately know its very clean. But for a lower SINAD score for a power amplifier you may want to look into the specifics of the amplifier. So say the amp I am interested in tested not so great on the SINAD scale but when looking at the data I see that noise is extremely low and the distortion product is mainly second harmonic then I can say the amp should work just fine with my efficient 105db horns because importantly noise is low which is very important for sensitive speakers and the distortion is 1% at full power but it's second harmonic dominant, well my speakers are efficient so I won't be near max power and so at listening levels I know I'll just have a little bit of second harmonic which we know to be rather benign.

 

When consumers get caught up with SINAD ratings I think they can lose perspective and miss out on gear that they may truly love but with a little insight and a better understanding of what the data is telling us we can gain a better picture of if the SINAD will matter for our application or not. It is complex and confusing, for a very good SINAD score it's safe to say the device performs well but for the gray areas of lower SINAD scores doesn't necessarily mean the device will sound bad, you'll just have to dive deeper into the data or demo the device. This is especially true if you are not one to care much about distortion and especially if you even prefer a little coloration which I do not deem as very important for an entertainment device, we aren't reading MRI scans or doing dialysis for Pete's sake. A recording engineer may want to keep their system as squeaky clean as possible to get an accurate mixing and mastering product but the rest of us end users it's all about entertainment and enjoying what comes into our ears.

  • Like 1
Link to comment
Share on other sites

One thing to keep in perpective (especially for non-engineer types) is that measurements are "tools". It seems a lot of people dismiss measurements and say their ears are the ultimate test & measurement system. Our ears can easily mislead us and test measurements help guide us to make informed decisions. It has been well established that measuremetns alone should not be used to formulate opinions about what equipment to buy; you should listen to music you are familiar with as well on the equipment. In addition, most people don't know how to interpret the test measurements which only adds to the confusion when trying to make an educated buying decision. 

 

Not in this case, but in a lot of other cases as captainbeefheart has pointed out, there is a huge amount of info on YouTube, but you need to wade through it and separate the good from the bad info. 

 

 

Link to comment
Share on other sites

Ahhhhh, everything old is new again.

 

To the tune of Star Wars Theme.....

 

"Spec wars, let's have more spec wars, nothing but spec wars, dah dah ta da......"

 

As I used to hear said in audio rooms across the landscape while sales guys and audiophiles alike would shake the manual and say "CAN YOU HEAR THEM SPECS?????" Not to mention that as illustrated above the KIND of distortion and where it rests along the chain is a HUGE deal. I've got tube amps run a half a percent THD in a lot of instances but the listener would never know it, especially when fed by teflon capped upstream signal that is supremely high bandwidth, ever exacting in its details and delicacy, yet a little THD in the final output stage vs a competitor amp with "better specs" isn't the point or necessarily indicative of a better result. Stuff like TIM and other problems are a much larger issue but oft get overlooked as does the topic of "tonality" which is really hard to measure, at least to the extent that we haven't come up with one yet, akin to "degrees Kelvin" that we use for lighting. My head imagines a similar infinite palette of choices/degrees if we had such a system/"spec" to measure that.

 

I'll recall our dearly departed brother Dave Mallette who stated "If it sounds good, it IS good!" and while the stuff beneath the hood and attention to detail does matter it's not always in the way that you think.

  • Like 1
Link to comment
Share on other sites

Here's a project I did a year ago.  The measurements are the most basic, 1kHz THD and bandwidth.  Without even listening to it, I'd state it will sound pretty good.  But, do the measurements tell the whole story?  No.

 

Swapping the 5AR4 rectifier for a solid-state plug in improves the low end a bit which I don't need a plot to tell me, I can hear it.  I can however measure the difference in 20Hz distortion and the temperature of two resistors.   The hotter resistors mean output tube life will be lowered.  A compromise I'm not willing to take.   Why?  Because the amp, as is, satisfies everything I look to a tube amp to satisfy.

 

CHA_CHB_BW.jpg

build_#2_1.jpg

chassis_17.jpg

CHA_CHB_1kHz_%THD.jpg

SN_003-1.jpg

  • Like 3
Link to comment
Share on other sites

Now companies are going to start clogging the arteries of components just so they measure well. Amazes me that beginners are arrogant enough to think they can judge components based on measurements they’re not properly trained to understand, myself included.

Link to comment
Share on other sites

On 4/16/2022 at 2:35 PM, captainbeefheart said:

I suppose it's no surprise because for whatever reason Bob decided to change the front end and reduce the amount of feedback which easily shows in the measurements.

I don't think Bob had anything to do with the design. The original Tubes4HiFi design was/is by Roy Mottram, and was put out a long time ago. I have one of the earlier designs, that uses 12at7s and doesn't has the CCS circuitry. Still in a small bin not put together, but it's getting moved up on my priority list.

 

Bob Latino may be out of the game now, as Roy has this posted on his website:

 

As of May 1st 2022 the amplifiers will be sold by my new partner Ray Longhitano,
with continued support by Roy Mottram (that's me) who started this business back in 1989.

 

  • Like 1
Link to comment
Share on other sites

On 4/18/2022 at 2:12 PM, captainbeefheart said:

If they really wanted to make the front end better they could have done a much better job in my opinion. They added a tube and got worse performance

Sort of added a tube. The original 7199 tube has a pentode and a triode in the same tube.

 

As for the bad layout, considering when it was originally designed, he may have done the layout with tape.

 

IIRC, Craig at NOSVALVES pretty much used the same circuit for his amps, but as in all things, I could be wrong.

Link to comment
Share on other sites

3 hours ago, Marvel said:

Sort of added a tube. The original 7199 tube has a pentode and a triode in the same tube.

 

The original used one triode and one pentode for the front end per channel. The newer design uses three triodes in the front end per channel, I should have stated it this way in that they added another active device in the front end per channel and it didn't improve the performance.

 

3 hours ago, Marvel said:

IIRC, Craig at NOSVALVES pretty much used the same circuit for his amps, but as in all things, I could be wrong.

 

I don't think I know him, is he a regular on here?

Link to comment
Share on other sites

On 4/24/2022 at 11:19 AM, Audible Nectar said:

Ahhhhh, everything old is new again.

 

To the tune of Star Wars Theme.....

 

"Spec wars, let's have more spec wars, nothing but spec wars, dah dah ta da......"

 

As I used to hear said in audio rooms across the landscape while sales guys and audiophiles alike would shake the manual and say "CAN YOU HEAR THEM SPECS?????" Not to mention that as illustrated above the KIND of distortion and where it rests along the chain is a HUGE deal. I've got tube amps run a half a percent THD in a lot of instances but the listener would never know it, especially when fed by teflon capped upstream signal that is supremely high bandwidth, ever exacting in its details and delicacy, yet a little THD in the final output stage vs a competitor amp with "better specs" isn't the point or necessarily indicative of a better result. Stuff like TIM and other problems are a much larger issue but oft get overlooked as does the topic of "tonality" which is really hard to measure, at least to the extent that we haven't come up with one yet, akin to "degrees Kelvin" that we use for lighting. My head imagines a similar infinite palette of choices/degrees if we had such a system/"spec" to measure that.

 

I'll recall our dearly departed brother Dave Mallette who stated "If it sounds good, it IS good!" and while the stuff beneath the hood and attention to detail does matter it's not always in the way that you think.

 

Thanks for your reply I really enjoyed it and I agree with you so I'll clarify my position so it accurately depicts my thoughts on "better" for this specific instance. I do not get hung up on .5% THD with a tube power amp especially if it's lower order harmonic distortion. The transducer will always add far more distortion especially at higher power levels so to me it's moot if the amplifier has .5%THD at 35 watts because the speaker will most likely be something like 8% THD especially at low frequencies, literally 30% THD isn't uncommon at lower frequencies with some speakers at higher output levels.

 

We have UL push pull EL34 output stage, when I look at the original vs the newer front end I can see the logic in that they feel a long tailed pair would be a better phase inverter, this can be true when the loads of the two phases are not equal; e.g. driven near 0v where the grid circuit acts like a reversed biased diode and starts to conduct clamping the voltage. But the amp shouldn't be driven into clipping like a guitar amplifier. The split load phase inverter used in the original is actually a very good phase inverter with great balance. The better option would be to just add one more bottle per channel, keep the split load but use the extra bottle as common cathode drivers to drive the power tubes, ala the Williamson circuit. Although the split is under unity gain, with the addition of the extra common cathode stages after the phase inverter you end up with more gain than with just one stage driving a long tailed pair. The original design is very elegant, the pentode/triode single bottle has a ton of gain in the pentode stage and then the triode section is a very accurate phase inverter. The load they use at the tail has frequency dependent loading problems, at high frequencies the tail is at a much lower impedance. It's just odd the new design has such high distortion at such a low power output, 1% at 10 watts, that's pretty terrible for a push pull EL34 UL amplifier as far as anyone is concerned. The newer design is a clone of the famous mullard circuit, even back then with no active load for the LTP tail they were getting .05% THD at 20 watts!!! Now as I mentioned I really don't get too hung up on vanishingly low distortion but being an engineer for a very long time you learn with little effort where certain designs should end up at in regard to distortion. So seeing a Mullard clone with poor specs tells me the circuit wasn't really designed very well because for what it is naturally the distortion should be much lower. Mind you I'm not saying all good amps have very low distortion, I have single ended pentode amps that have 1% THD at 12 watts so that's pretty bad a push pull amp performs worse when it already has a huge advantage of second harmonic cancellation in the output stage and the fact my single ended amp is second harmonic dominant. So my SEP amp with 1% THD at 12 watts will sound much cleaner because it's second harmonic dominant where the new design of the ST70 is third harmonic dominant, but second harmonic is quite high for a push pull amp telling me the front end is not very linear. Overall I just don't think they did a very good job with what they had to work with.

Link to comment
Share on other sites

6 hours ago, captainbeefheart said:

 

The original used one triode and one pentode for the front end per channel. The newer design uses three triodes in the front end per channel, I should have stated it this way in that they added another active device in the front end per channel and it didn't improve the performance.

I wonder if that one Amir bought was built correctly and set up right. A lot of folks have liked Roy's boards over the years.

 

The 7199 was designed for audio, but was also out of production for quite a while.

 

We'll see how mine sounds when I get it done. No exotic parts... I think the manual/build instructions have the schematic. If it does, I will send it to you.

 

6 hours ago, captainbeefheart said:

I don't think I know him, is he a regular on here?

He was on here all the time. He marketed  his own amps, the VRDs.

 

http://nosvalves.com/

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...