30W 6vDC transformerless PSU

[watchmaker]

I have a pre boombox Sony stereo FM radio I really like.  AC/DC in theory except the special AC cord is AWOL; old cheater that also pushes contacts to switch from AC to 6VDC.

4 D cells are perfect.  So I decided to all an external DC power port.  HAH!  Anything but a linear PSU screws up the performance.  I do not understand the issue, but I just want to make a Transformerless adapter that converts 120 VAC to DC so that I can then drop it down to 6 VDC with 3 to 5 Amps.

I have looked at transformerless schematics but they seem to be all for under a couple hundred mA.  I looked at the AC-DC IC datasheets and they seem to not provide the needed power output.

Is what I want possible without a transformer?

THANKS

Dewey

[Zero999]

The short answer is no. Don't use a transformerless power supply. It's not practical at this power level and is unsafe because it isn't isolated and your boom box doesn't have mains rated insulation.

If a switched mode power supply causes problems then it's due to lack of filtering or leakage through the Y capacitor.

[Konkedout]

The short answer is no. Don't use a transformerless power supply. It's not practical at this power level and is unsafe because it isn't isolated and your boom box doesn't have mains rated insulation.

If a switched mode power supply causes problems then it's due to lack of filtering or leakage through the Y capacitor.

I agree...DO NOT USE TRANSFORMERLESS.  That is dangerous.  "lack of filtering"...The AC-DC Power adapter is probably a switching power supply which is creating radio frequency noise that interferes with your radio.  Quiet switching power supplies do exist, but the easiest way to avoid the problem is to find a linear power supply which uses a mains frequency (50/60 Hz) transformer.

One possibility may be to find a good common mode choke to connect between a switching power supply adapter and your radio.  This is not guaranteed but may be smaller/lighter/cheaper than a linear power supply.

[watchmaker]

Thank you both.  I kinda figured.

Have no idea why the radio hates the switching power supplies.  AFC goes to Hell and distortion.  Even with a constant current adjustable adapter.  I have to raise that to 9VDC  (can go to 24 VDC) just to get anything.  And 4 D cells work for at least 30 hours.

Go figure.

[PCB.Wiz]

  I have to raise that to 9VDC  (can go to 24 VDC) just to get anything.  And 4 D cells work for at least 30 hours..

If it can tolerate 6-24v, you could look for low noise linear power supplies.
Not that cheap, but the audiophiles love them.

[Andy Chee]

I would test your radio using an old AT/ATX computer supply connected to the 5V rail.

An AT/ATX supply is obviously not a permanent solution, however it should prove that the radio is capable of good performance using switchmode.

However, my solution would be to replace the AC input with a C8 figure-8 socket, and install an isolation diode to isolate the D-batteries whilst on mains power.

[watchmaker]

WIBF'd!

After trying an L7806 and a LM317 I decided it was NOT the volatage supply.  I hypothesized that my 6 inches of 22 AWG created a voltage drop.  MEasured at PSU: 6.12 VDC.  Measured at connection to radio; 3.2 VDC.

Changed to 18 AWG wire and it works just fine with the 6 v wall wart I tried originally.

Live and Learn (if you can live long enough).

[langwadt]

Thank you both.  I kinda figured.

Have no idea why the radio hates the switching power supplies.  AFC goes to Hell and distortion.  Even with a constant current adjustable adapter.  I have to raise that to 9VDC  (can go to 24 VDC) just to get anything.  And 4 D cells work for at least 30 hours.

Go figure.

3-5A and 30 hours on D cells doesn't add up

[NiHaoMike]

3-5A and 30 hours on D cells doesn't add up

That might be the peak draw with the volume turned all the way up. Under more reasonable conditions, it will be much less.

[watchmaker]

The issue was the resistance of the 22AWG wires used to connect to the various PSUs acted as a voltage divider; cutting the voltage in 1/2.  Before I figured this out, the only thing I could think of was I needed a 1 A PSU; and since I was going to do that, might as be 4 or 5A.

Early on I missed a clue.  Using an adj 24 V power brick, I needed to increase the output to 12 VDC in order to get function.

As it is, I learned much regarding transformerless PSUs, several voltage regulators, buck converters and most importantly, the impedance of the load vs the resistance of the supply wires.

Was very surprised that such a short run could result in cutting the supply voltage in half.

Now at normal volume it draws 200 mA at 6 V with both the radio and MPX unit (the second speaker) which is below the max rating of 3.5 W. Very interesting old school design, 1966 Sony 8f-38 with STA 38 w. Excellent reception 70 miles from Boston and great sound for WCRB. (ears are over 71 and we did not know about PPE).

Now I have a LM317 set for 6.2V from a 12 V supply, and a LM7806 on a heat sink, ready to go.  Plus both datasheets.

[m k]

So it's not so "transformerless" anymore.
The idea of linear is to have a power reserve, first from capacitor and then from transformer.
The idea of switch mode is to save expenses, you're giving that away.

If linear has a current reserve you have a voltage reserve, sort of.
So you are constantly "overloading" the switch mode.
Nothing wrong with that, power reserve is a power reserve.
But your current power usage for Sony is a bit less than 30W.

Couple questions.
Which one is more economical, this or linear?
What if you double the current power usage?

[Konkedout]

I agree.  Those numbers do not add up.

A foot of AWG 22 wire should be OK for 2 Amps of current.  (0.011 ohms per foot if I remember correctly).  2 Amp of current would kill D alkaline batteries in 5-10 hours.

I am thinking that your AWG 22 wire has some defect causing resistance WAAAAAAAY more than it ought to.

[Zero999]

He didn't say how long the cable is. It also might be steel, rather than copper. It's possible the supply decoupling capacitors inside the radio have dried up making amplifier and other circuitry unstable, with a high impedance power supply.

[watchmaker]

Had I been less annoyed and more inquisitive, I would have/should have checked the resistance form to battery box to the connector.

I always check connections when I make a lead/cable assembly, but I do not remember checking the connection after I soldered it to the battery compartment.

It likely was a cold solder joint at the battery compartment.

Oh well.  Still, it was a great opportunity to learn all sorts of other things.  Problems are just opportunities to learn.

Edit missing word