ST-70 Custom Died - Where To Start?

musichal

poet emeritus
Perhaps I just found why I shouldn't own a tube amp. Music was playing, at low-ish volume. My wife was standing beside the amp and I was looking at her, in conversation. There was no flash, no pop or buzz or noise of any kind. No odors. The music quit playing, the tubes faded. Won't fire back up. Should I even try replacing any tubes? Or just try to find a tech? I own a multimeter, and there ends my knowledge base.

A rustnrot (Tom Lang) build, following info provided:

Here is an amp I recently completed. Everything on this amp is new except for
reusing original Dynaco A470 output transformers. This amp uses a Classic Valve
Design "Dynamull" board http://www.classicvalve.ca/docs/DynaMull_docs.pdf
However, the input 6EJ7 tube is connected as a triode rather than a pentode, in
my opinion, even better sounding.
For a chassis, I did it differently this time so as to have the inputs, outputs,
and power connections on the back panel.
Individual bias pots are accessible from the top and test points for each tube
are accessible from the top adjacent to each output EL34 tube.
This amp also uses a MUCH more capable power transformer and also utilizes solid
state power supply rectification for increased capacity. The capacitor boards
are of my own design and use surface mount high voltage ultrafast rectifiers.
I get continuous output power of 45 watts per channel RMS with both channels
being driven simultaneously...
Detailed instructions for setting the
output tube bias will also be included as well as a schematic diagram.


The DynaMull driver board is a Classic Valve Design adaption of the
Mullard 5-20 circuit for the Dynaco ST-70 with some modernization.
Two main differences are the choice of input tube and the Constant
Current Source (CCS) in the tail of the phase inverter.
Why the EF184/6EJ7? It is a fantastic tube with clean sonics, low noise
(brand dependent) and a lot of transconductance.... significantly more than the
EF86. It is also cheaper and easier to find.
The CCS in the tail of the 6FQ/CG7 is a "no brainer" - superior
performance AC and DC. A time-tested cascode design that we and others have
used successfully for years. Proximity to the 6FQ/CG7 keeps temperature
nearly constant for minimal temperature coefficient drift.
We have attempted to keep to the classic design overall, as it already is
one great sounding circuit.

upload_2016-12-6_17-22-45.pngupload_2016-12-6_17-22-45.png


Assembling the PCB is fairly straightforward and needs no adjustments or
alignment when complete.
Numbering of the off-board connections is the same as the original ST-70
board for simplicity. The only difference is the grid bias resistors for the EL-34's
are each brought out to their own pad should you wish to use our ST-70 power
supply board (or your own circuit) for balancing of the power tube bias
(instructions for such provided on the ST-70 power supply board
documentation). Should you use the original ST-70 bias circuit, which only
provides bias for left and right channels, tie the two pads labeled "21" together,
as well as the pads labeled "6" together and run a wire to their respective bias.
Image for this shown later.
Due to the density of the board, certain components must be installed
before others. Here is the procedure (components mounted on the top side of
the board unless noted):
- Install C101 and C201 on bottom side of the board
- Install R103 and R203
- Install R115, R116 and R215, R216
- Install R110, R111 and R210, R211
- Install C105 and C205
- Install C103, C104 and C203, C204
- Install R105 and R205
- Install D101, D102 and D201, D202
- Attach heat sinks and insulators (if used) on Q102 and Q202
- Install Q101, Q201 and Q102, Q202
- Install R113, R114 and R213, R214 on bottom side of the board
- Install remaining top-of-board components.
- Install remaining bottom-of-board capacitors
- Install sockets
C107, C108 and C207, C208 require "spaghetti" on parts of their leads to
keep from shorting out on some components, if you have used small pin-spaced
capacitors (we used Panasonic ECW, which are small). Suggestion is shown
below.
The wire "spaghetti" can be a piece of insulation stripped off a wire with
600V rating, actual radio-TV spaghetti, or even 1/16" walled polyurethane
pneumatic tubing available from a commercial supply house (the volts/mil
breakdown of polyurethane is very high - just don't use it on components that'll
get hotter than ~70C.).
While we are on the subject of smaller or larger pad spacing components
(often capacitors), there is a right and wrong way of mounting them for reliability.
Wiring
Now that your board has the components mounted, you need to prepare
the filament wiring.
Each socket has unnumbered pads for filament connections. The
6FQ/CG7 tubes also have a pad on pin 9. The pad on pin 9 of the 6FQ/CG7
tubes can have jumpers to ground. Some brands of 6FQ/CG7 tubes have a
shield between the sections that is to be grounded. Other brands have no shield
and the pin floats free internally. Better to ground this pad than let it float (we
didn't do this on the PCB because some experimenters like adding a small
negative voltage here).
The heater pads (pins 4 and 5) of the left 6FQ/CG7 and 6EJ7 need to be
tied together with a twisted pair of wire (to reduce hum). Same for the right.
Each will go to their respective power supply leads on the ST-70 power supply.
The image below shows this, as well as the bias connection we talked
about earlier.

Note for Advanced Builders: You can use a pair of 3EJ7 for the input tubes
and wire their heaters in series. Tie the now 6V string to either channel.
To aid builders with the pad-number-to-wire arrangement, here is the
original Dynaco ST-70 schematic:

upload_2016-12-6_17-27-34.pngupload_2016-12-6_17-22-45.png upload_2016-12-6_17-27-34.png


Pad numbers "3" and "18" are not used on this board and the 18K/2W
resistor may be added externally to the preamp power sockets, if desired.
Pad number "19" is labeled "B+1" on the DynaMull board and pad number
"20" is labeled "B+2".


The power supply of the original ST-70 requires two resistor changes:
If you are using our ST-70 power supply board, use the components
shown, 47K, 2W and 4.7K for R9 and R10 on that board, respectively.
If you have a different B+ off point "C" than the stock Dynaco due to the
use of

upload_2016-12-6_17-29-48.png


If you are using our ST-70 power supply board, use the components
shown, 47K, 2W and 4.7K for R9 and R10 on that board, respectively.
If you have a different B+ off point "C" than the stock Dynaco due to the
use of solid state rectification or a different power transformer, please adjust the
4.7K resistor so the B+2 is in the area of 415V to 420V.upload_2016-12-6_17-22-45.png
 
The fade you speak of would seem to suggest a bad rectifier. Easy fix if so - just make sure you retention the socket when replacing ANY tube.

Be even easier if you were to re-tension ALL the octal sockets and it started working again.
 
If it is the rectifier is bad do you have an old 5u4? Could work for a quick check, I wouldnt run one all the time in this amp. No Money might have something there too. How is the wall socket?
 
Note he said in the description that this has solid state diodes, and no rectifier, and that the "tubes faded"

If this meant filaments stopped glowing, then it's either the fuse, the switch, the line cord, or the power transformer.

WHY did the fuse blow? Maybe they used the original size fuse, and the new power transformer is too big for it? Maybe the peak current of the new filter caps is too high?

Maybe it was just a bad fuse?

I would suggest the following - check the fuse, and replace if bad. If the amp works normal, just keep using it or change to the next largest fuse size.

If it blows right away, then dig into it farther.
 
Has a new solid-looking heavy duty IEC power cable that has shown no evidence of problems. Fuse must be internal, and I will need the wife's assistance getting to that due to my weakness, so it may have to wait for the weekend. Good thing I have spare amps. If I had an old 5u4 (tube?) in my hand I wouldn't know it. Was hopeful that the IEC plug was not fully inserted at the amp for a moment, but that was not the case.
 
Original Dynaco has one of those black fuse holders accessible from the outside - are you sure it's internal?

If it's just a computer style power cord easy enough to try another one.
 
Note he said in the description that this has solid state diodes, and no rectifier, and that the "tubes faded"

If this meant filaments stopped glowing, then it's either the fuse, the switch, the line cord, or the power transformer.

WHY did the fuse blow? Maybe they used the original size fuse, and the new power transformer is too big for it? Maybe the peak current of the new filter caps is too high?

Maybe it was just a bad fuse?

I would suggest the following - check the fuse, and replace if bad. If the amp works normal, just keep using it or change to the next largest fuse size.

If it blows right away, then dig into it farther.

Sounds like the place to start, when I can get at the fuse, and will check switch then, also. The switch, like everything else except the OPTs, is brand new and heavy duty. The chassis is new and the faceplate is blank stainless, everything is on the rear panel.
 
Has a new solid-looking heavy duty IEC power cable that has shown no evidence of problems. Fuse must be internal, and I will need the wife's assistance getting to that due to my weakness, so it may have to wait for the weekend. Good thing I have spare amps. If I had an old 5u4 (tube?) in my hand I wouldn't know it. Was hopeful that the IEC plug was not fully inserted at the amp for a moment, but that was not the case.
I was looking at the drawing and noticed a rectifier tube. That if went bad could be replaced. Mine uses a 5ar4. But a 5u4 will work but is a hard start compared to the 5ar4. I just finished reading that you seem to have a SS a rectifier. I
have several 5u4s i use in a pinch until I find what rectifier tube i need.
 
If there is an extra bit of space above where the IEC plug goes in, the fuse may live in there. Some of them have a fuse holder as part of the power entrance socket setup.

sk21a.jpg
 
Before you open it.

1. Take multimeter and set to ohms, touch probes together. Should read close to 0

2. Unplug IEC cord from back of amp, touch probes to hot and neutral blades of IEC plug on back of amp (two farthest apart). With amp turned on should read a few ohms. if it's reading open, you have a blown internal fuse. With switch off should read open.

3. If not reading open, repeat test with cord attached, touching proves to blades of plug. Ignore the grounding prong.

4. Open amp and check switch, internal fuse, and transformer for continuity.

5. Take photos of inside of amp and post here, I think with a multimeter and collective AK knowledge we can get this thing going.
 
If the fuse is blown and it's a normal fast blow fuse I'd try replacing it with a slow blow. But since it died while in use, not on startup, I'm guessing something else failed. Surface mount solid state rectifiers?? Hmmm.
 
Back
Top Bottom