[BUILD] STA LEVEL PCB OFFICIAL BUILD & HELP THREAD

[bernbrue]

Well done! Very nice build!  8)  ;D
Bernd

 

[ilfungo]

Hi Bernd
Wich fuse value  for EU operation?
Thanks!

 

[bernbrue]

Hi,
Original manual says 2A for 115V operation. I would try a 1A.
Bernd

 

[djfatum]

The first ready  8)

https://youtu.be/YSgwCePI6BE

My DIY projects

 

[bernbrue]

Very nice! I really like the round meters. And soon in stereo, whow!
Cheers
Bernd

 

[ilfungo]

Hi Bernd
original H pad + 1k pot  need to be terminated with 600 ohm resistor?
Thanks!!!

 

[bernbrue]

Not really needed.
Bernd

 

[ilfungo]

Finally with standard pad (10db+1k pot) + another 6 db fixed H pad  I'm in the right spot for my work!
UTC 0-1 input
Edcor Output

Noise is incredible low for a tube comp!!!
Sound is icredible!
Thanks  bernbrue Great project!!!

Some samples - wetransfer:
https://we.tl/qwIat9x6Lv


Thanks!!!

 

[bernbrue]

Thank you. You're welcome!
Bernd

 

[j.frad]

Hi guys!
I finished building my sta level a few weeks back and I'm test driving it.

It compresses nicely but sounds a bit too fuzzy too quickly.
It's fantastic for bass, but I can't see myself tracking vocals through it with a risk of sudden distortion.
It was my understanding that unless using too short release times, it should not sound distorted.

Do you have a clue of what it could be?

I am also using an edcor input transformer until I know the unit works perfectly (then I'll invest in sowter).
I had bad surprises using edcor transformers (took them out of my daoc because of distortion, wouldn't work in my NYD EQ...)
I know the input transformer shouldn't be influenced by how I drive the unit but what is you experience using an edcor input transformer?

thanks!

 

[bernbrue]

Hi,
You may want to read the chapter "In use" carefully.

http://www.soundonsound.com/reviews/retro-instruments-sta-level

The faster the time constants the more likely the unit starts to distort especially with bass heavy signals. I had very good results with Edcor transformers on the output. I wouldn't use an Edcor input transformer in this circuit.
Best
Bernd

 

[techineer]

Are there any specifications for this version, or are they the same as the original Gates unit?

I've been reworking a build that was done by one of the members of this group, and after getting it (apparently) working, it seems to require a very high level input signal before compression occurs, resulting in a very high level output. 

I'm wondering if it really is working properly, or if something else needs to be diagnosed.

 

[techineer]

In post #69, Bernd recommended a switch to ground in the F- line.  I am assuming from what I'm reading (correct me if I'm wrong) that the F- line should normally be connected to ground.  If you use a pushbutton type switch, be sure to use one that is NORMALLY CLOSED and is rated to carry the filament current.  Such a switch is usually described as ON/(OFF) in the catalogs.

 

[techineer]

Beginning around message #220 and possibly some others is mention of a problem in the +5V supply.  This unit was sent to me to iron out the bugs, and I see what is probably a design flaw in the 5V supply. 

The secondary winding for this supply is 5VAC (measures 5.6VAC RMS) After rectification it should measure up to 6.7VDC (or less, depending on load)  (5VRMSx1.414 (peak) -1.2V (Diode drops)).  The LM317 requires at least 3 volts greater input voltage than the desired output voltage in order to maintain regulation, or more as the load current increases.  http://www.ti.com/lit/ds/symlink/lm317.pdf (See page 9.) 

In order to have a regulated 5 volts output, the DC input voltage must be at least 8 volts and preferably more.  The AC voltage required for this is at least 6.5VAC.  ([5v out +3v (reg I/O diff) +1.2V (diode drops)]/1.414 =9.2v/1.414=6.5VAC rms)  I would suggest another 9V winding instead of a 5V one. 

You could also jumper the 9VAC input from the filament circuit over to the 5VAC input, as there should be some excess capacity in that winding.  The tube filament current ratings add up to 2.1A, while the transformer winding is rated at 4A.  The 5V regulator is limited to 1A, so that should be usable.

 

[bernbrue]

Are there any specifications for this version, or are they the same as the original Gates unit?

I've been reworking a build that was done by one of the members of this group, and after getting it (apparently) working, it seems to require a very high level input signal before compression occurs, resulting in a very high level output. 

I'm wondering if it really is working properly, or if something else needs to be diagnosed.

I need to know if the input pad is installed, which input transformer was used and what the voltages are at the dedicated test points. The other aspects you mentioned are basically right or have already been discussed before.
Best
Bernd

 

[dmp]

Beginning around message #220 and possibly some others is mention of a problem in the +5V supply.  This unit was sent to me to iron out the bugs, and I see what is probably a design flaw in the 5V supply.
The secondary winding for this supply is 5VAC (measures 5.6VAC RMS) After rectification it should measure up to 6.7VDC (or less, depending on load)  (5VRMSx1.414 (peak) -1.2V (Diode drops)).  The LM317 requires at least 3 volts greater input voltage than the desired output voltage in order to maintain regulation, or more as the load current increases.  http://www.ti.com/lit/ds/symlink/lm317.pdf (See page 9.) 
In order to have a regulated 5 volts output, the DC input voltage must be at least 8 volts and preferably more.  The AC voltage required for this is at least 6.5VAC.  ([5v out +3v (reg I/O diff) +1.2V (diode drops)]/1.414 =9.2v/1.414=6.5VAC rms)  I would suggest another 9V winding instead of a 5V one.
You could also jumper the 9VAC input from the filament circuit over to the 5VAC input, as there should be some excess capacity in that winding.  The tube filament current ratings add up to 2.1A, while the transformer winding is rated at 4A.  The 5V regulator is limited to 1A, so that should be usable.

The V+ rail is only used for the relays, which should be 5v versions if you are setting up a 5v V+ rail. Usually relays will still work with slightly less voltage. And I expect the LM317 will delver pretty close to 5v. I would suggest measuring your v+ voltage with the relays engaged and off. If you have anything else on the v+ rail you could move it to the 9v or, if it is a bulb, the 5v AC. But I expect it should work with 5v AC in and 5v relays.  The higher the supplied voltage to a rectifier, the more heat you have to dissipate to keep the rectifier happy.

 

[bernbrue]

Are there any specifications for this version, or are they the same as the original Gates unit?

I've been reworking a build that was done by one of the members of this group, and after getting it (apparently) working, it seems to require a very high level input signal before compression occurs, resulting in a very high level output. 

I'm wondering if it really is working properly, or if something else needs to be diagnosed.

You will find a PDF file with specs in the third post. In the second post you'll find further information on F- to ground.


10) Calibration:  make shure that heater (f-) is connected with a wire  to chassis ground. Make this connection switchable by using a momentary switch.  As soon as the reference to ground is interrupted the unit starts to hum. Turn the trim pot between the two EL90s until hum is reduced to a minimum. That's it. See attached file.

Bernd

 

[frederickalonso]

Thinking to make one in the future.

 

[gareth33]

I've started the build, I have a tube noob question, which gauge hookup wire should I be using? Voltage rating and so on? Some parts will need larger gauge than others?

I look forward to having it up and running, have gone for Sowter input and output and Don Audio power.

Thanks for any advice.

Gareth

 

[bernbrue]

Hi Gareth,
I always recycle wire from old computer psu and other electronic junk. For the molex connectors I take preconfigured connector with wires.

e.g. http://de.manu-systems.com/images/L/DEV-5-WAY-MOLEX.jpg

Bernd