1
\$\begingroup\$

I am currently making a tube amplifier for a school project. I started out with a finished design, but have changed a lot and need some guidance as a result.

I am currently using a full wave rectifier bridge I made with 1N5404 diodes, and I will be rectifying 50Hz 230V AC into 325V DC. I have a 470uF capacitor I intend to use to smooth out the voltage, and I am wondering if it's even safe to put it directly after the bridge without resistors in series to limit the surge. The diodes should withstand a 200A surge, but I’m not sure how big the surge will get. Is the isolating transformer going to get saturated and limit the surge enough? I haven't bought one yet, but it will be a 30-50VA transformer. Also, would it be safe to test the circuit without an isolating transformer? I would like to test it out before I get it!

\$\endgroup\$
0

2 Answers 2

Reset to default
2
\$\begingroup\$

Also, would it be safe to test the circuit without an isolating transformer?

You can burn your finger on a hot resistor or tear your skin on a sharp edge of a PCB. Nothing is safe; it's all a matter of reducing risk so I'd use an isolating transformer and there's no reason to believe it will saturate due to the load in fact, it'll saturate less on full load than on no-load (watch this space for anyone disagreeing!!).

The leakage inductance of the transformer windings will tend to offer a little bit of current limiting for each time the diodes re-charge the capacitor.

I will say this about the diodes - they are rated at 400V and I believe you might expect diode failure when other circuits switch in and out around where you plug this thing in - I'd go for something a bit bigger like 1N5405 (rated at 500V in reverse). I'd also ensure the cap is 450V rated for the same reasons of reliability.

\$\endgroup\$
5
  • \$\begingroup\$ Thanks for the advice! My capacitors are indeed 450V rated. About the diodes; I'm thinking of the initial surge when the capacitor gets 325V DC and acts as a short circuit for a brief moment. I expect everything to run smoothly after that when the capacitor is fully charged, as the circuit only draws 0,1A at a pretty constant rate. \$\endgroup\$ Commented May 11, 2014 at 20:19
  • 1
    \$\begingroup\$ I'd consider an inrush NTC thermistor such as described here: en.wikipedia.org/wiki/Inrush_current_limiter - they have an initial series resistance that is high enough to stop tens of amps flowing and when warmed up (in a matter of a second or so) their resistance drops to allow bigger currents through. \$\endgroup\$ Commented May 11, 2014 at 21:27
  • \$\begingroup\$ I agree with Andy here, the NTC is an extremely common and really simple way of stopping inrush currents to large capacitor banks on input power circuits like this - it often is used to prevent 5-10A fuses from blowing on start-up of the device/supply. On a side note, Andy is pretty boss and knows what he's doing. we should all be glad he hangs around this forum to divulge his wisdom to those who need it so much! \$\endgroup\$ Commented May 12, 2014 at 0:05
  • \$\begingroup\$ Using an isolation transformer when hooking up test equipment is a good idea, but if a final device (which won't have any human-touchable parts) is supposed to work without an isolation transformer it's important to keep in mind that a product may work perfectly on an isolation transformer but die rather dramatically when plugged into "raw" mains. Using a switch some distance away from the product to apply mains voltage can be a good idea. \$\endgroup\$ Commented Mar 4, 2016 at 15:42
  • 1
    \$\begingroup\$ Diodes have an inrush specification for just this initial surge when the circuit is first switched on. The 1N540x series is rated 3A continuous, 200A for one mains half-cycle. You won't need much stray series impedance to keep the current to that level. As supercat hints at, the extra impedance of an isolating transformer may make the difference between keeping to the inrush spec or not, so test it without at some stage. \$\endgroup\$ Commented Jun 28, 2016 at 5:10
2
\$\begingroup\$

Tube amp builders generally use 1N4007 (or UF4007 if switching noise becomes an issue) which has Vr of 1000V. They're so cheap there's no reason not to use them.

Some rules about tube amps:

  • ALWAYS use a power transformer. Never plug directly into the wall.
  • ALWAYS use a 3 prong plug and properly earth ground the chassis.
  • ALWAYS use a fuse.

The last thing you want is to pick up your guitar and be directly connected to the line voltage. It makes for a very unpleasant day, whether you survive it or not.

I highly recommend reading the "Safety Tips for Working on Tube Amplifiers" at Aiken amps.

Incidentally, one way to do cheap isolation transformers is to take two 12V or 6V transformers and connect the secondaries together. This will give you line level voltage and isolation. I've even used transformers taken from $1 wall warts that I pick up at thrift stores. Just make sure it can handle the current. You can use the 6V or 12V secondaries to create your filament supply.

\$\endgroup\$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.