SMPS Adjustable DC Power supply 0/10 - 300 volt/0.3-0.5 Amp

[pikstart]

Hi.

I 'd like to upgrade my design of such a DC voltage power supply that currently, to reduce power losses on the pass element, uses SCR controlled rectifier. The SCR is fired when the difference between the output and smoothing cap is less than about 10 volts.

This method has two main disadvantages.
1. Needs a rather bulky transformer to isolate main power.
2. Produces nasty pulses (of about 0.1 to 5 Hz) on the rectifier that are also audible on the transformer itself as ticks that user notice it.
3. Those rapid pulses will eventually reduce the life of smoothing capacitor.

I 'd prefer total SMPS solution with a smaller ferrite transformer, but the wide adjustable output is something that could not find how to do it.

Any ideas welcome.
Thank you
Ioannis

 

[rjenkinsgb]

How about using phase control, with the phase proportional to the pre-regulator <> output regulator difference?
That should keep the current well down, as there would be no irregular top-up pulses.

Also be sure to use both half cycles from the transformer; using only one or the other magnetises the core and reduces efficiency.

 

[Papabravo]

The description of what you have now is rather imprecise. What I'm hearing is that you have an offline power supply that converts AC line voltage to DC at up to 300V. What I don't know is the secondary voltage on the mains transformer. If it is too low and you want to make a boost converter you will be working with a large boost ratio, in excess of 4-5, which is impractical. If it is closer to 300 VAC then a non-inverting buck-boost topology might work. Lastly you could employ a second transformer to make a forward converter, or a flyback converter. Depending on your particular circumstances there may be other topological possibilities.

A schematic of the existing design would be most helpful -- neatly hand drawn if that is all you have.

 

[pikstart]

Thank you for the replies.
Phase control is not really too different of what now the circuit is doing. It still needs the heavy and expensive transformer to work.

The current design is a classic linear power supply. The main "improvement" is that instead of a classic rectifier to convert the secondary 220 vac to about 300 vdc, it uses two SCRs and the trigger is controlled by the difference of the smoothing capacitor and the output voltage of the regulator. Whenever this difference is greater than say 10 volts, the SCRs are triggered and charge the capacitor with a pulse.

If necessary I may isolate that part of the circuit, but prefer not to (yet).

My preference would be something like this:
Line input, rectifier and about 320 Vdc on the cap. Then, a buck converter to regulate the output voltage from say 0 or 10 volts min to 300 volts max, at 500mA max. Can accept also less current if this is too much, at about 300mA. Obviously I will add short circuit protection along with current leakage to protect the user and the power supply.

The wide span, I understand, a bit difficult to obtain in SMPS.

Ioannis

 

[Papabravo]

The main problem with using a buck converter is having the duty cycle vary from 10/320 = 3.23% to 300/320 = 93.75%. Most buck converters will have a problem using a duty cycle over 50%. At the higher duty cycles you may have to tolerate increased current ripple and potentially sub-harmonic oscillation. Neither of these things will be good for the output.

 

[Papabravo]

As you can see from the following simulation there is only a narrow range of duty cycles that can control the output effectively. There may be a way around this, but the method escapes me at the moment. The duty cycles from 20% to 80% all result in approximately the same voltage. and 4% gets you about 220 Volts. I have no idea how to get 10V with some level of stability. It would take an incredibly small duty cycle, where very small changes in that duty cycle will result in very large large changes to the output voltage.

 

[pikstart]

I was afraid of that in the low and high DC.

Initially, I tested the P.S. with classic linear design and a great deal of power loss. After testing the operation of the rest circuit, I stepped in the SCR control of the rectification with great reduction in losses, about 5-8 watts (depending on output voltage) and with active cooling at low rpm of the fan.

But would really make the difference with a new design based on SMPS. Then got stuck in the wide range of DC that needs to control the output voltage. Such designs do exist with output voltages 10-300 up to 20-6000 volts:

Mini basic electrophoresis apparatus, GEP-PH-300

1. Reliable work and stable performance HT-300 basic electrophoresis apparatus Adopt advanced microcomputer control power supply design scheme. Good stability, light weight and small heat dissipation. Effectively improve service life. 2. Beautiful appearance, fastening and durability The latest...

bioevopeak.com

Multi-Functional Electrophoresis Power Supply, GEP-PW-W600T - Bioevopeak

Output range voltage:5-600V；current:1-1200mA；1-500w Sensitive limit:voltage 1V, current:1mA, power:1w Output: constant voltage, constant current and constant power

bioevopeak.com

Electrophoresis Apparatus, GEP-PH-ECP3000

Features: LCA display with backlight (128*64 pixels) Microprocessor intelligent control has: Overload, no-load, leakage and other protection functions It can store 10 commonly used electrophoresis programs Automatic memory function Standard running function Automatic judgment function Constant...

bioevopeak.com

All seem to have switching power supply as the enclosure size is small enough.

Thanks again,
Ioannis

 

[Nigel Goodwin]

How about a normal transformer type SMPSU, with various output taps on the transformer, with automatic selection of different taps as the voltage is altered?. A common solution in lab PSU's - including the one I'm sat next to.

 

[pikstart]

Well, the point is to avoid the classic, bulky, expensive transformer with the smaller, light and cheaper ferrite one, that is driven by a switching element (bipolar or mosfet).

Currently I do have in my design a toroid 150VA transformer that is powering the supply and the rest logic and control circuits.

 

[Nigel Goodwin]

Well, the point is to avoid the classic, bulky, expensive transformer with the smaller, light and cheaper ferrite one, that is driven by a switching element (bipolar or mosfet).

Currently I do have in my design a toroid 150VA transformer that is powering the supply and the rest logic and control circuits.

Which was why I suggested a transformer type SMPSU, which uses a light and small ferrite transformer.

 

[pikstart]

OK, I am not familiar with such type of core transformer for line power connection. But the rest of the problems remain, like the SCR controled rectifier, the strong pulses that are audible and do reduce capacitor life.

 

[Nigel Goodwin]

OK, I am not familiar with such type of core transformer for line power connection.

I've never made any such suggestion - if you don't understand what an SMPSU is you're going to struggle with your project.

A switch-mode power supply generally works by converting the AC mains to DC, it then 'chops' the DC to form high frequency AC, this is then fed through a high frequency (ferrite cored) to produce an isolated output on secondary windings. This is rectified, to produce DC, and the voltage is measured, with a correction signal send back to the primary side to alter the mark/space ratio of the 'chopping' to regulate the output.

This is how almost all electronics in your house works, TV's, DVD's, phone chargers, computers - almost anything you can think of - it's been that way for decades.

 

[pikstart]

We are lost in the translation...

Thanks Nigel. That is what I was after from the begining. I'd like to get rid of the classic toroid and move on to a more advanced switching design.

The problem, as Papabravo noted, is that the duty cycle needs to range from 3 to almost 100%. This is difficult to obtain and also maintain stable regulation of the output voltage.

Ioannis

 

[Nigel Goodwin]

We are lost in the translation...

Thanks Nigel. That is what I was after from the begining. I'd like to get rid of the classic toroid and move on to a more advanced switching design.

The problem, as Papabravo noted, is that the duty cycle needs to range from 3 to almost 100%. This is difficult to obtain and also maintain stable regulation of the output voltage.

Ioannis

Which is why I suggested multiple taps on the output of the transformer, that auto-switch as you alter the voltage - one or two taps should be enough?.

 

[ChrisP58]

I have a SMPS based power supply on my bench that almost does what you're asking. Mine is a Sorensen DCS33-33 and only goes from 0 - 33 volts, but there are other members of the series that go to higher voltages, up to 600 volts.

It uses a ferrite transformer driven by a full bridge power stage. That topology can operate from zero to near 100% duty cycle.

Page 3-2 of this document shows the block diagram of the supply.
https://www.programmablepower.com/products/dc-power-supplies/a6e490d1-1a8f-41c5-b8ee-d3badb4afe95 Bad link. See post 17 below.

 

[Papabravo]

I have a SMPS based power supply on my bench that almost does what you're asking. Mine is a Sorensen DCS33-33 and only goes from 0 - 33 volts, but there are other members of the series that go to higher voltages, up to 600 volts.

It uses a ferrite transformer driven by a full bridge power stage. That topology can operate from zero to near 100% duty cycle.

Page 3-2 of this document shows the block diagram of the supply.
https://www.programmablepower.com/products/dc-power-supplies/a6e490d1-1a8f-41c5-b8ee-d3badb4afe95

Your link appears to be broken. I've seen a full bridge used in an inverter, but not in a DC to DC scheme. I don't have a clear picture of what you are describing. I know -- words fail us sometimes. A block diagram might be helpful.

 

[ChrisP58]

Sorry about the link. I guess the page isn't formatted for link sharing.

Try the "Operations Manual 1 /1.2 kW (rev L)" link on this page.
https://www.programmablepower.com/products/dc-power-supplies/dcs-series

In power supplies 'full bridge' means that both ends of the load (transformer primary) are alternately switched to both power supply rails.

 

[Papabravo]

Sorry about the link. I guess the page isn't formatted for link sharing.

Try the "Operations Manual 1 /1.2 kW (rev L)" link on this page.
https://www.programmablepower.com/products/dc-power-supplies/dcs-series

In power supplies 'full bridge' means that both ends of the load (transformer primary) are alternately switched to both power supply rails.

Great. Now all the TS hast to do is come up with the scratch to buy one. Problem solved.

 

[ChrisP58]

Great. Now all the TS hast to do is come up with the scratch to buy one. Problem solved.

My apologies. It wasn't my intention to say "buy this"

I was only showing that there is a working power supply topology that can be used to build what he's looking for.

 

[Papabravo]

Well, the TS has essentially two options:

1. He can try to buy what he needs
2. He can try to design and build what he needs.

Trying to design and build what he needs might be a major challenge, it might even cost as much as a purchase after running down and procuring the components, fabricating a PCB, and building the enclosure. This does not seem like a weekend project, but a major engineering undertaking.

After looking at the schematic/block diagram in the link I can tell you that the full bridge driver is very similar to the circuit you might find in an inverter. Even so that multiple winding transformer may be a challenge to engineer. Designing transformers is not exactly my area of expertise, but I know some people who could probably pull it off.