Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need an unregulated (or input-tracking) DC-DC voltage inverter

Status
Not open for further replies.
Hi All,
I'm pretty new to switching power supplies, but what I want is a switching DC-DC converter that inverts its input supply.
The specs I am aiming for are as follows:

*Vin Max = 24V
*Iout Max ~= >0.8A
*Vout = -(Vin)
*Should work down to a few volts, but this is not critical
*Continuous short circuit capable (and generally robust)
*Efficiency and size is not of particular interest

I have successfully used several DC-DC converter ICs (LMZ14201, TL2575-ADJ and MC34063A) in inverting mode and each of these devices can satisfy my requirements above, however they are all regulated, so whilst I may adjust the output voltage with no issues, I cannot think of a way for the device to "track" or mirror whatever the input voltage is.

For comparison purposes, the MAX7660(S) is unregulated and would be great, but it's limited to 13V max and about 100mA.

I have noticed that the datasheets show the feedback pin being compared to a voltage reference. I have experimented with varying voltages on the feedback pins of several devices but since the comparator will be outputting either "voltage too high" or "voltage too low", I can't think of how I might be able to achieve a circuit where Vout is just -(Vin).

If anyone has any suggestions, I would be most grateful.


For your reference (devices I have right now): TL2575-ADJ, LMZ14201, MC34063A.


PS: I've searched some manufacturers to see if a part is available, but nothing that meets my desired specs (namely output current) seems to be available. If anyone knows of a part (preferably Ti / Intersil / Microchip / Linear) I would also love to hear it.

Thank you!
 
Why not use an appropriate kind of FET in place of one of the feedback resistors. The gate could go (via an appropriate circuit) to the supply voltage, so its resistance varies with supply. It won't be accurate tracking, but at least it will be tracking, I think.
 
Deleted
 
Last edited:
If it was a non-inverting SMPS like a 24v->12v buck SMPS you can make it track simply by using an opamp to compare the output voltage to the input voltage (in a ratio if needed) using the 2 opamp inputs, and the output of the opamp goes to the 1.2v Vsense pin of the SMPS IC.

You should be able to do a similar thing with an inverting SMPS although you need to deal with using +/- voltages.
 
How about a transformer design?
You could use an SG3525 or TL494 or similar controller and feed a push-pull transformer with something like 1:1.2 ratio, then recitfy the center-tapped secondary side. If you don´t care about the size and efficiency of the thing, then just get some (any :D) core and a magnet wire and try it out. And without feedback it of course will be tracking the input voltage.

Continouse short capable means that it has to keep struggling, or can it turn off on overload?
 
Last edited:
Take a look briefly at this application note:

https://www.electro-tech-online.com/custompdfs/2012/11/CD00003773.pdf

The form of a power supply you really want, I think you want is a tracking voltage regulator. These are typically used to get a ground that is 1/2 way between the total voltage.

There is plenty of reasons why it's would be difficult to mirror an input voltage exactly, but to mirror a setpoint should not be too hard.

A "tracking" regulator is different than say applying a +setpoint to one input and an inverted setpoint to the other.
 
Thanks for the replies, everyone.
I'm sorry to respond so late, (busy at work) but I thought I would share the solution I ended up developing. I'm sure it's flawed but it seems to work reasonably well.

A schematic says a thousand words so I've linked it, as well as the LTSpice sim file. If you want to simulate it, I've linked the LM337 & MC34063A libraries that you'll need.

Since I've found the MC34063A to be very prone to dying, I might change it for the TL2575-ADJ (which I found to be quite robust and good-performing) if I need to (the design makes this trivial).

Anyway, I just wanted to post this for completeness, however any further comments would be great.

Thanks!

PS: Sorry about the external links, the file attaching system gives me an error at the moment.
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top