Dynaco Mark III?
- Thread starter Blackdog61
- Start date
albusark
Dynaco Fan
The Mark III is my favorite amp setup of all time, I like it better than McIntosh or any other. For good reasons I think - the circuit is elegantly simple. It's easy to work on or restore, it changes character based on the tubes you put in, and it's extremely reliable in my experience. I have two sets I've restored and they're my personal end game amps, I adore them and think Hafler made something very special in designing them. If you're not experienced with them, take a look at the inside chassis and see how few components are in the signal path.
That said, the ST-70 is a quieter amplifier with different characteristics. A pair of Mark III's fully restored original circuit with KT88's are far more powerful than an ST-70 at max volume. I have no clue why anyone would say different.
If you don't want to bother your neighbors and want a gold standard hi-fi goosebumps experience, the ST-70 is a fantastic choice. But if you want a louder, more immersive tube amp or maybe want to try modern low efficiency speakers then a pair of Mark III's is great choice. Also, it could simply be that I love how KT88's sound.
That said, the ST-70 is a quieter amplifier with different characteristics. A pair of Mark III's fully restored original circuit with KT88's are far more powerful than an ST-70 at max volume. I have no clue why anyone would say different.
If you don't want to bother your neighbors and want a gold standard hi-fi goosebumps experience, the ST-70 is a fantastic choice. But if you want a louder, more immersive tube amp or maybe want to try modern low efficiency speakers then a pair of Mark III's is great choice. Also, it could simply be that I love how KT88's sound.
Lavane
Lunatic Member
60 watts and up starts getting heavy. The PPP KT-88 mono blocks I have are no fun at all to move around. I've been wanting to wrap wood all around the chassis of those amps but every time I look at them I put it off.
Over the years have owned 4 or 5 st-70s and 4 pairs of MKIIIs. ST70 is a pretty nice sounding amp. You can get a better sounding amps, but likely will be a big step up in cost vs. the ST70. I would be more concerned with the front end source or upgrading the preamp. Those two options may give you a bigger step up in resolution than going from a ST-70 to MKIIIs.
If not for the fact I had been after an example of this specific amp for years I wouldn't have bought it just because I don't want stereo amps that big. Its a Bogen DS-265, not exactly something that comes up very often.
The writer of "Lilienthal 100 Amplifiers" said many years ago improvement could be made in the front end of the ST-70. After doing some research on my part I found the designer of the tube4hifi VTA driver board made some of those recommendations. But then on some youtube videos comparing the two some preferred the original ST70 over the so called improved VTA circuit. Much the same testing the two amplifiers. In some ways the original test better than the VTA driver board one.
Last edited:
my single biggest complaint with the various aftermarket boards is that test data showing how it actually performs is usually either completely absent or very incomplete. The original has loads of testing from when it was in current production, and into the modern era that confirms the old tests were pretty accurate. I get that numbers aren't everything but I still want the technical data.
Lavane
Lunatic Member
The biggest reason I went with Bob Latino's VTA board when I bought the amp kit was the fact that 12AU7's and even 6SN7's were still in production by various companies. The existing supply of good 7199's used in the Dynaco ST-70 is getting kind of thin. Heck, they weren't in production anymore when Dynaco started using them. There are a couple other tubes that can be used, but again, they arent made anymore either. I buy mostly new production tubes just because I don't like paying huge inflated prices for old stock tubes.
Over the years I've listened to plenty of Dynaco ST-70's.
I like Bob Latino's VTA amps sound. I could care less about the CCS, but I do like the use of commonly found tubes. Some have stated the VTA amps sound like a solid state amp, whatever that means.
SS PP amplifiers do share some of the same characteristics of a tube PP amplifier only tubes do it better. Not so much the BJT ones but the class A mosfet SS PP amplifiers. If you look at the curves of mosfets and pentodes you will see much the same curves in the data sheets. A constant current running in the tube no matter the voltage is a good thing for sure.
Except that never happens. With an audio signal the current has to slide up and down the load line along with the voltage changing. Otherwise no audio amplification. The only time the current is a near constant is at idle with no signal.
Thats a solid reason. There are other triode/pentodes but I don't think any of those are in current production either so eventually it will just become a problem again.
Talking about two different things. Always an advantage having a constant current being supplied to the tube. Of course the current must change when an ac signal from the grid changes the current or you will not have amplification. You really do not want the current running in the tube changing without a signal and that is what happens without a CCS due to AC voltage supplying the amplifier is constantly changing. A CCS circuit is a plus in an amplifier.
Not trying to be a smart ass, there are many sharper knives in this kitchen than myself.
My understanding of how a tube operates is the grid is nothing more than a variable switch changing the current running in the tube. That is all it does, change the current using the small AC signal. A small negative voltage is supplied to the grid limiting the current flow somewhat where the AC signal can vary the current flow running inside the tube. The current is changed to voltage in the load, most time a resistor. No voltage gain in the tube itself just current gain. That is my understanding how a tube operates.
Just trying to pass along what little I have learned in my 77 years playing with electronics. If I am wrong I am sure I will be corrected.
Adding a CCS is something we as diy'ers can do to improve a circuit. Manufacturers do not often use it because of cost and complexity adding another circuit. Keeping the cost of their product down to be competitive. Different story in high end audio products.
Just trying to help.
My understanding of how a tube operates is the grid is nothing more than a variable switch changing the current running in the tube. That is all it does, change the current using the small AC signal. A small negative voltage is supplied to the grid limiting the current flow somewhat where the AC signal can vary the current flow running inside the tube. The current is changed to voltage in the load, most time a resistor. No voltage gain in the tube itself just current gain. That is my understanding how a tube operates.
Just trying to pass along what little I have learned in my 77 years playing with electronics. If I am wrong I am sure I will be corrected.
Adding a CCS is something we as diy'ers can do to improve a circuit. Manufacturers do not often use it because of cost and complexity adding another circuit. Keeping the cost of their product down to be competitive. Different story in high end audio products.
Just trying to help.
in a diff amp it actually is damn close to constant current at the cathode. Consider the outputs are inverted so as one triode draws more current, the other draws an equal amount less. Its not completely perfect since tubes are not perfectly linear or perfectly matched but its as close as you're gonna get provided it stays within the normal area of the curves.
A CCS does a better job of keeping it perfectly constant but a sufficiently large "tail" resistor does pretty good for a completely passive device. The imperfection can usually be accounted for with one plate resistor being a bit smaller than the other. Less parts to install, less parts to buy, less parts to fail. One of these days I'll put a CCS in something but honestly so far everything I've built with an LTP / diff pair inverter has done it the old school way, and honestly thats because I had the parts to do it that way and didn't feel like buying parts to do it the other way.
A CCS does a better job of keeping it perfectly constant but a sufficiently large "tail" resistor does pretty good for a completely passive device. The imperfection can usually be accounted for with one plate resistor being a bit smaller than the other. Less parts to install, less parts to buy, less parts to fail. One of these days I'll put a CCS in something but honestly so far everything I've built with an LTP / diff pair inverter has done it the old school way, and honestly thats because I had the parts to do it that way and didn't feel like buying parts to do it the other way.
Lavane
Lunatic Member
That was my thought exactly. I wouldn't have bought the ST-35 amp kit from Dynakit, or an original Dynaco for that matter if the 12DW7 wasn't still in production.
I switched to a SS rectifier in my ST-70 mainly because I don't really trust new production 5AR4's after loosing two, and I refuse to pay $$ for a more reliable NOS 5AR4 thats still going to be marginal in that amp anyway. And no way I'm paying the big bucks for a NOS GZ-37's for my ST-70, or my mono blocks.
Many have installed a couple of 1N4007 diodes across the rectifier tube that way keeping the tube sound and protecting the tube at the same time. Never tried it myself but planning to on my next build. I believe it is called the yellow paper trick or something like that. Just install the diodes on the socket as if there is not a tube there. Not exactly sure how it effects the voltage drop across the tube which I am trying to keep.
I always say I like tube amps because the parts aren't obsolete, unlike decades-newer solid state amps but there are certainly exceptions to that. 7199 specifically and other triode-pentodes are on that list. I'm not super concerned about that, plenty of 6U8 and similar around, but were I looking to mass produce an amplifier it wouldn't be using those.
I'm up to 7 5AR4s in use currently, and I really don't know how I got here because its not somewhere I planned to go. Three of them are modern and seem to be OK, one I'm pretty sure would be fine with modern, but the big Bogen uses 3 and I don't know about that. One I'm certain would be fine since it just runs the screens and small signal tubes, the parallel pair running the output tube plates I don't know about though. Higher voltage and current in that spot. Timing is also a minor worry, if the screen supply comes up slower its no problem, but if it goes hot before the plates do its hard on the output tubes
The diodes go between the transformer and anodes of the tube. Basically it adds some peak inverse voltage protection, but does nothing for the forward current. I've tried it and got some data, it insignificantly affects voltage regulation and output voltage drops by about 1 volt.
if you want to read my testing on it. Post 44 is also relevant since it gives the as-built performance to compare.
Lavane
Lunatic Member
Are there any issue's that could arise running the 5AR4's, or any rectifier tube for that matter in parallel? Say one comes up a little faster or slower than the other, or if one becomes weaker that the other?
I glanced over the article. Are you saying you did not find a significant difference in voltage drop using the yellow sheet mod with the 2 diodes? I am curious because I am building the Antek ST-70 EI kit and I am going to use the rectifier tube for appearance and for the voltage drop of the tube but want to install the diodes to protect the tube but do not want to increase the voltage.The diodes go between the transformer and anodes of the tube. Basically it adds some peak inverse voltage protection, but does nothing for the forward current. I've tried it and got some data, it insignificantly affects voltage regulation and output voltage drops by about 1 volt.
if you want to read my testing on it. Post 44 is also relevant since it gives the as-built performance to compare.