Another PCB check - PSU with Transformers

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I think that looks pretty good Dan. 
One thing I noticed, and it may be that the resolution is too fuzzy to tell?  But on the TL783 to the left:  Does the bottom trace to the middle pin have enough clearance from the pin above?  Maybe bring it out a bit first before scooting it back? 
That's all I have to say 🙂
 
BluegrassDan said:
And 3D.

Thanks, everyone, for all the help. I'm new to PCB design. This is just my second board.

You should have seen my second board.  You should see some of my current boards for that matter! 
😊
 
BluegrassDan said:
One more "FINAL" version. Needed to clean up the silkcreen and add two more mounting holes halfway.

Do you really need C1, C12, and C (I don't see the number) across the secondaries?
R9 or R10 is redundant.
What's dissipation of D6? I't is IMO too small and too close to C9.

You can do that board on bottom layer only. Than use top layer to beef traces and make eventually ground planes.
 
R9 and R10 are two voltage dropping resistors, dividing 66k @4W between two 33k @ 2W resistors.

D6 is a GBU802. 8W, 200V. https://www.mouser.com/ProductDetail/Diodes-Incorporated/GBU802?qs=sGAEpiMZZMtQ8nqTKtFS%2FJolQOfQMTyVY01irg1iCuA%3D
 
BluegrassDan said:
R9 and R10 are two voltage dropping resistors, dividing 66k @4W between two 33k @ 2W resistors.
You can't replace 66k/4W resistor with two 33K/2W. Anyway you don't need so low values of these resistors just for heater bias.
D6 is a GBU802. 8W, 200V. https://www.mouser.com/ProductDetail/Diodes-Incorporated/GBU802?qs=sGAEpiMZZMtQ8nqTKtFS%2FJolQOfQMTyVY01irg1iCuA%3D
It looks good but I would put a small heatsink on it.
 
BluegrassDan said:
Please help me understand ground planes. How would it help this design?

A (ground) plane is a concept taken from RF circuitry where it is used to isolate tracks (and get better stability) by shunting parasitic capacitors to the ground.  There is it possible because the input/output impedances are usually low, 50 or 75 ohms.
In audio spectrum it is used mainly to get lower impedance/resistance of common  signal distribution lines as ground, power tracks etc. If they are correctly implemented, they are good for EMI shielding too. In your case, you can implement three separate grounds,  which help to have lower hum or other problems, as Audio1Man already noticed. 
 
BluegrassDan said:
Am I undermisstanding? Two 33k 2W in series cannot replace a single 66k 4W?
Yes, they can. My bad. I didn't decoded this part of the board correctly. Could you post the schematic of the whole power supply? 
 
moamps said:
You can't replace 66k/4W resistor with two 33K/2W. Anyway you don't need so low values of these resistors just for heater bias. It looks good but I would put a small heatsink on it.
Many tubes specify no more than about 20K resistance between heater and cathode. 66K and 22K give 25% HT heater elevation.

Cheers

IAn
 
ruffrecords said:
Many tubes specify no more than about 20K resistance between heater and cathode. 66K and 22K give 25% HT heater elevation.

Does this resistance should have the same value if one or more tubes are connected to the one heater power supply? What happens if this resistor is shunted with a capacitor? Is resistance Rhk always equal to Rh-ground?

Btw:
(Designing High-Fidelity Valve Preamps by Merlin Blencowe)

 
 

Attachments

  • heater-cathode.png
    heater-cathode.png
    107.1 KB · Views: 11
moamps said:
Does this resistance should have the same value if one or more tubes are connected to the one heater power supply? What happens if this resistor is shunted with a capacitor? Is resistance Rhk always equal to Rh-ground?
The resistor value is the same no matter how many tubes, The bottom resistor is always decoupled by an electrolytic so that at ac the heater and cathode are connected directly together.
Btw:
(Designing High-Fidelity Valve Preamps by Merlin Blencowe)

I am not sure where Merlin gets his 150K value from. I used to use a much higher values - something like 300K at the top and 100K at the bottom. But then I came across a number of data sheets where it was stated it should not be more than 20K so I changed the design. There is no obvious reason to make the resistor values high other than the small HT current of about 3.4mA at 300V HT. It is an unfortunate fact that the vast majority of data sheets do not specify this parameter so I choose to err on the side of caution.

In my latest designs I have reduced the HT voltage from 300V to 250V. With the tubes I use I can now dispense with heater elevation altogether. Which is handy because I can now use the heater supply for LEDs and relays instead of having to provide a completely separate supply for this purpose.

Cheers

Ian
 
Back
Top