Variable DC HV Power Supply

Thread Starter

Wendy

Joined Mar 24, 2008
21,914
@Dzoro @bertus One of the frequent threads we moderators frequently have to shut down is transformerless power supplies to create high DC voltages. I have decided to show one way to do this that meets AAC TOS and COD ( Terms of Service and Code of Conduct .

555.png

Parts List

Qty 1 U1 7555 CMOS Timer IC
Qty 1 U2 LM317 3Terminal variable voltage Regulator IC
Qty 1 T1 Transformer 120VAC to 12VAC center tap optional
Qty 4 D1-D4 1N4006 Diodes A Diode Bridge with suitable specs can be substituted
Qty 1 Q1 TIP 102 NPN Power Darlington Transistor
Qty 1 Q2 TIP 107 PNP Power Darlington Transistor
Qty 1 C1 .01µF 16VDC (or higher) Capacitor
Qty 2 C2,C9 0.1µF 16VDC Capacitor
Qty 2 C3,C4 10µF 16VDC Electrolytic Capacitor
Qty 1 C6 1000µf 16VDC Electrolytic Capacitor
Qty 1 C7 100µf 300VDC (or higher) Electrolytic Capacitor
Qty 1 C8 100µF 16VDC Electrolytic Capacitor
Qty 1 R1 1KΩ ¼W 5% Resistor
Qty 1 R2 1.2MΩ ¼W 5% Resistor
Qty 4 R3,R4,R5,R6 100KΩ ¼W 5% Resistor
Qty 1 R7 120Ω ¼W 5% Resistor
Qty 1 R8 1KΩ trimmer variable Resistor
Qty 1 R9 36Ω ¼W 5% Resistor

Warning This supply creates Dangerous voltages and lethal currents. If you do not treat it with caution it can kill you, and it will hurt the entire time you are dying. Neither All about Circuits or the author assumes liability for this circuit, Build at your own risk!


Now that the warning is out of the way, I would rate this circuit suitable for intermediate to advanced users. CMOS chips (the 7555) can be easily zapped with static electricity, so ESD precautions are advised. This circuit is as of yet untested. but I have high confidence it will work as is.
The voltage generated is not regulated, When loaded it may drop substantially, I will draw a regulated version in the near future. Since the transistors will be operating in their linear region heat sinking will be required.
The voltage can be increased even more by grounding the center tap of the transformer It is up to the user to adjust the diodes and capacitor C7 specs to compensate the added strain on these components.

Theory of Operation


I have always felt the conventional 555 was a remarkable IC. The CMOS version (such as the 7555 addresses the 555 weaknesses and adds a few more. Among the things it improves on is a much wider range of power supply voltages (typically 2VDC to 18 VDC)and a true rail to rail output. The AC level is a function of the power supply voltage fed to the 7555.This comes at a price as it can only drive microamps at the lower supply voltages. For this reason I made the transistor drivers as high impedance as I could. The back to back capacitors allow polarized electrolytics to be used, as they are much easier to acquire than large unpolarized electrolytic capacitors. The transistor circuit drive circuits will only handle the AC signal, Both the 555 and755 are fairly stable in frequency over their power supply ranges. I chose 60 Hz because that is what most USA transformers are designed for if there is a request for it I will calculate the values for 50Hz and our EU friends.

I spent quite a while coming up withe the resistor values for U2, it will adjust1.625VDC to 10.8V. This will cover the voltages a user might need.
 
Last edited:

Thread Starter

Wendy

Joined Mar 24, 2008
21,914
fixed. This is a work in progress I will be updating this post continuously for a couple of days.
 

wayneh

Joined Sep 9, 2010
16,128
Question: Why the coupling capacitors, particularly C5 and C6, in series with opposing polarities? Does this ensure they never see reverse polarization?
 

Thread Starter

Wendy

Joined Mar 24, 2008
21,914
It is a cheap and easy way of making unpolarized electrolytic capacitors. The signal will be true AC when it is all said and done. They make unpolarized electrolytics , but they can be hard to find and tend to be more expensive.
 

wayneh

Joined Sep 9, 2010
16,128
It is a cheap and easy way of making unpolarized electrolytic capacitors. The signal will be true AC when it is all said and done. They make unpolarized electrolytics , but they can be hard to find and tend to be more expensive.
I don't think that arrangement accomplishes what you think it does. In simulation, I see damaging reverse voltages appearing on C5. I'd love to be proven wrong.
 

Thread Starter

Wendy

Joined Mar 24, 2008
21,914
It is a very old trick that works.The forward biased cap takes on all the strain. I have used this technique for decades.
 

Sensacell

Joined Jun 19, 2012
2,517
Capacitors back to back with diodes parallel to limit the reverse volts the caps see to ~ 0.7v?

Seems like a more bulletproof approach?
 

Kjeldgaard

Joined Apr 7, 2016
381
A few words more about the C5 and C6 pair.

There should not be a need for a bipolar capacitor at the point in the circuit.

The primary winding of the transformer will be 0 Volt DC and the Q1 / Q2 amplifier output will always have a positive DC voltage.
 

wayneh

Joined Sep 9, 2010
16,128
A few words more about the C5 and C6 pair.

There should not be a need for a bipolar capacitor at the point in the circuit.

The primary winding of the transformer will be 0 Volt DC and the Q1 / Q2 amplifier output will always have a positive DC voltage.
That’s what I see, and in the circuit as drawn that puts an average of -1/4Vcc on C5. C6 sees the other 1/4 of Vcc but in the right orientation.

I recently built an inverter for an EL project and it’s similar to this project. I used only a ”C6”, not two in series as we see here. The circuit performs just as the simulation predicted.

[update] - Confirming @Wendy 's use of this trick, I found a discussion of the topic here: https://electronics.stackexchange.com/questions/21928/can-you-make-a-non-polar-electrolytic-capacitor-out-of-two-regular-electrolytic
It clearly works, and it relies on an interesting property of electrolytic capacitors. The simulation doesn't account for this property. But here the trick is not needed anyway.
 
Last edited:

Thread Starter

Wendy

Joined Mar 24, 2008
21,914
You will see an AC voltage across C6. Back to back Electrolytic caps work, I have a lifetime of experience that says otherwise, It is also why I do not trust simulators. People have presented over unity circuits based on simulators. If you truly do not trust this approach you will need to find non polarized electrolytics to replace them. Odds are they are just two caps back to back internally anyhow.
 

DickCappels

Joined Aug 21, 2008
6,005
You see pulsating DC that never drops below ground.
I would have expected to see AC, especially if the transformer were non-ideal. Why do you think it would be pulsating DC? Just curious.

Edit: I see. If she reversed the polarities of the two capacitors it would be AC to some extent. I guess the point is that there should be little or no DC current flowing, which is what the capacitors care about.

It seems that as a refinement she could eliminate C5 since the driving signal has a large positive DC component and the load is grounded.
 

wayneh

Joined Sep 9, 2010
16,128
It seems that as a refinement she could eliminate C5 since the driving signal has a large positive DC component and the load is grounded.
Exactly. There's no mechanism to cause a reverse voltage on C6. You get double the capacitance by eliminating a component.
 
Top