|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
70.184.213.18
We recently modded our fifth Dynaco ST-70. It's a customer unit, so we were paid to install a factory-built board, do our mod to the power supply, unscrew some stuff someone else had done, supply the 6SN7s, bias and test.
This board had two upgrades: the Russian paper-in-oil coupling caps, and increased capacitance on the input board. It also has the later, larger power transformer.
It doesn't have our 500uF second filter cap, but 150uF in capacitance for the second cap. Still has 50uF of input capacitance.
It has customer-supplied JJ output tubes.
This is clearly the best-sounding ST-70 we've done. It sounds...neutral. Most importantly to me, it sounds not like a tubed amp, not like a solid state amp, just...right. Excellent dynamics, excellent separation of instruments.
So what's the difference between this amp and the others we've done? We're going to try to find out. The first avenue of inquiry is the russian K40y capacitors. We're going to change the caps in another ST-70 we have to K40Ys and see how it sounds.
I believe we're going to stop putting 500uF caps in the second stage and stick with 150 - 200uF there. It should be easier on the power transformer (slightly greater conduction angle) and since the bass on the current amp is so good (and un tube-like) I doubt that there will be any degradation in bass.
Previously I was really proud of our (and tubes4hifi's) mod, but thought the McIntosh 240 was still a better-sounding amp. Now I'd put the ST-70 up against it. It will take some investigation to discover which of the differences take it from really good to truly great.
But I will tell you that the lowly ST-70 can be a truly great amp. And this isn't commercial, so please mods, don't delete. We use the 6SN7 tubes4hifi board (no affiliation other than customer), and we beef up the power supply. There's our whole recipe, open source.
Follow Ups:
that just because they still use the transformers, the chassis, and other various hardware of the original Dynaco that it is a modded Dynaco. And for the most part it is but it isn't in the real world. You start changing the circuitry for whatever reason, maybe for the better, or maybe not, it isn't what Hafler intended so therefore it isn't a Dynaco anymore.
Just my thoughts!
,
I have almost always found that too much is just as bad as too little when it comes to filter caps.
I can't see how we can 'hear' a VR, except via voltage drop.
All rectifier stages have a job, low ripple DC. We can't hear 0 Hz.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
What is good with the ST70The output stage uses very good output transformers.(Good)
The output tubes are EL34 in push pull with fixed bias. (Good)What is bad with the ST70
The power supply caps are not very good (Bad)
The coupling caps are not very good (Bad)
The driver circuit is marginal at best (Bad)
The circuit board and sockets are old and crusty (Bad)
The power transformer is too small (Bad)
The B+ Supply is extremely stressed (Bad)
Pulling 250mA out of a single 5AR4 into a 50uF cap is cruel, adding another 500 uf after a silly tiny choke is pretty criminal.So, when you are done changing the bad, what you have has almost nothing to do with an ST70 other than the output transformers (which are quite nice)
If you want to make it better, fix the B+ circuit in its entirety to be able to supply 400mA of current. Put it in a properly sized chassis with a double rectifier and you will have something very good. It just has no relation to an ST70.
Edits: 01/23/16
it's Hafler's attempt to provide the most bang for the buck, and overall, succeeds pretty well. The power supply, while perhaps wimpy, does use good engineering principles and is about as good as you can get for the money. The driver may be the most compromised area, but once again a tried and true circuit that does indeed do its job pretty well. I'm sure that David would find quite acceptable a number of upgrades/mods as long as they addressed the shortcomings that the low price imposed upon the design, as long as those upgrades/mods did indeed make a better product, not a different product...
Mine certainly isn't much of an ST-70. The driver and power supply are from Curcio, the output tubes are 6B4G, cathode biased. The rectifier tube is a 5U4GT. No negative feedback.
The only original parts are the transformers, the chassis, the speaker screw terminals (no, I've never replaced them), and the power cord may be original, or it may be a crappy replacement. Pretty sure the power switch is original. There may be a few wiring posts in there that are original, I don't remember at this point.
Nope, not an ST-70.
I read what you posted ans I have to say I am in full agreement but not from any experience of knowledge(you guys are much better than me) but what others have said. A good friend and tech told me when you add too much UF's you change the T/C's(time constants) which can relate to what some would say too bass heavy. He suggests never going over two times the original but he also has said to use as little UF's as possible just to get it quiet. He always felt that when too much UF's are used the musicality is compromised. He also said that Hafler did not really match things to the schematic but he did stay in the tolerance %.
Also, many mods of adding this, that, or the other thing never really deals with the weakest link in the B+ chain. The power transformer. It is too anemic to deliver the current and have enough headroom to spare.
Whenever I did mods to an ST-70 or a Mk III I took out the choke, put a resistor(100-500 ohms or so) in its place, and used another decoupling stage(R/C) before the input stage. The amp was just as quiet and seemed to get faster(more neutral).
One thing I never realized was adding another tube rectifier but with that you would need a larger filament supply for it which again results in changing the power transformer.
With all the stuff you said to do it is not a Dynaco anything any more and it just pays to build your own amp from scratch.
The output stage is lightly modified: There is individual bias and the bias is lowered from 50mA to 40mA.
I think the important point is that for a few hundred dollars an ST-70 can become a great amp. The fact that one is fixing weaknesses in the original design makes it no less of an amp.
Mmmm Histoical
Maybe the O/P trafos can be saved. Everything else goes to the smelter!
Eli D.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: