 |
 |
| View previous topic :: View next topic |
| Author |
Message |
Mc
Joined: 07 May 2004
Posts: 11
|
| [off topic] 5V step-up DC DC converter? |
 Posted: Tue May 25, 2004 3:00 pm |
 |
|
Hi guys!
I guess this is kinda off-topic but anyway...
I'm building a PIC based microbot (1x18f252 + several 16f818 slaves) and now I'm in the bussiness of powering it. I was planning to use three or four 1.2V NiMH AAA batteries and some IC to stabilize the voltage up to 5V. I need around 1A output load...
I'd like to use a switching power supply, step up DC-DC converter given the high efficiency of these in order to maximize battery life, but I'm open to suggestions. Does any of you have any experience with these? How do you usually power up your PIC's? 
These days I've been playing with a MAX608 from Maxim (it seems ideal for my needs) but I can't get it to work, I guess it's for improper mosfet selection (I've tried both IRFD014 & IRL520N with no luck, though they seem to meet all requirements  ...)
Any suggestion? 
Thx a lot!!  |
|
 |
Guest
|
|
| Posted: Tue May 25, 2004 3:13 pm |
|
|
| Try TI or National. They have cheap low power DC-DC converters. |
|
|
SteveS
Joined: 27 Oct 2003
Posts: 126
|
|
| Posted: Tue May 25, 2004 3:22 pm |
|
|
How much voltage variation can your circuit tolerate? NiMh have a pretty flat discharge characteristic, so you may not benefit from regulation if you can tolerate an end point of 4.4V or so. You'd have to watch the initial voltage as it may be 5.6V or so. No regulator is the most efficient! Just so you think thru all possibilites.
If you need regulation then a switcher is a good solution. Try National's PowerBench. It let's you design a DC-DC based on your needs. Power supplies are not as simple as most think they should be. Switchers can be picky about layout and exact parts used.
- SteveS |
|
|
cbarberis
Joined: 01 Oct 2003
Posts: 172
Location: Punta Gorda, Florida USA
|
|
| Posted: Tue May 25, 2004 3:28 pm |
|
|
| Unless you are using other logic that is not as supply tolerant as a PIC you really do not need to step up or regulate your 4.8V (4 X 1.2V), Now, if you are using just 3 AA cells, you can still reliably run your PIC controller, as most PIC's will work OK up to 10 MHz with 3.6V. However if you are interfacing with other 5V logic, then I would suggest you use a boost type regulator and convert your battery voltage to a regulated 5V level. |
|
|
Mc
Joined: 07 May 2004
Posts: 11
|
|
| Posted: Tue May 25, 2004 3:46 pm |
|
|
Hi all
Well, yes, I need 5V regulated supply as I'm using some other components requiring 5±0.2V, I guess it's no other choice...
I'm now taking a look at National's LM3478, it's quite similiar to the MAX608 I'm trying to use so I think I'll get into the same design troubles ... I'll try to guess what's wrong with it...
Thx for the answers! |
|
|
ninjanick
Joined: 25 May 2004
Posts: 25
|
|
| Posted: Tue May 25, 2004 4:00 pm |
|
|
| Just a quick note, but Maxim has long lead times on some of their parts. Not sure if that's a concern, but it might. |
|
|
Guest
|
| Re: [off topic] 5V step-up DC DC converter? |
| Posted: Tue May 25, 2004 8:57 pm |
|
|
| Mc wrote: | Hi guys!
I was planning to use three or four 1.2V NiMH AAA batteries and some IC to stabilize the voltage up to 5V. I need around 1A output load...
|
First, I think your robot is not going to do much work for veyr long !
Secondary cells like AAA NiMH do not have much power !
You say 5V @ 1Amp ... 5 Watts !
AAA NiMH come at about 800mA/Hour wich menas you better plan on less that 45 minutes service life !
If you could use AA then it improves a lot. AA NiHM come at about 2300mA/Hour which could give you about 1h 45 minutes. !
I'm using a neat Maxim DC-DC converter it comes in a very small package and outputs stable 5V ( good enough for PIC projects) with 1.1V to 5V input.
It will continue to operate down to 0.7V input, but requires 1.1 to guarantee startup.
If you could use two primary C cells and this Maxim conveter then you are looking at about 7 hours usage, because it will drain the cells to 0.35V(each) and stop ! In other words way beyond the rate life of (example) Alkaline cells. NiMH will also work, but they tend to drop off very quickly and so there is nothing to gain by this feature. Either way the chip will work with both types.
Check out MAX1674EUA, it has o/p regulation +- 0.2V much like a cheap 7805 and can deliver 1 Amp @ 5Volts.
It requires a small 22uH to 47uH inductor (Digikey) and two 47uF tantalum caps. It also has a built in Low Battery detection circuit, which requires two resistors.
If you do use it, you MUST do a real PCB, this chips cannot be handwired, it too small. If you layout the PCB well, you can make a 5V 1Amp DC-DC on the area of a nickel coin.
I buy them from Maxim direct in lots of 24, because in lots of 25 and up, there is a 6 weeks lead time... !
Good luck
Hans W |
|
|
Mc
Joined: 07 May 2004
Posts: 11
|
|
| Posted: Wed May 26, 2004 3:34 am |
|
|
Hi!
Well, 1A is the maximum peak load I would need with everything turned on, in normal operation I guess it consumes no more than 300/400mA...
...but, ok, you're right, it's no problem to use AA or any other kind of batteries (9V...) as the PCB isn't still designed.
I'm testing the MAX608 in a protoboard with a couple of alkaline batteries just to make sure it works prior to making the PCB but this is the output I get:
... the mosfet tends to get warm so I think it starts to work in linear mode which is not adecuate  . I really don't know what's wrong with it...
Ah, I've just taken a look at the MAX1674 but I can't see it can deliver 1A, it seems to be designed for a limit of 0.5A |
|
|
Ttelmah
Guest
|
|
| Posted: Wed May 26, 2004 4:11 am |
|
|
| Mc wrote: | Hi!
Well, 1A is the maximum peak load I would need with everything turned on, in normal operation I guess it consumes no more than 300/400mA...
...but, ok, you're right, it's no problem to use AA or any other kind of batteries (9V...) as the PCB isn't still designed.
I'm testing the MAX608 in a protoboard with a couple of alkaline batteries just to make sure it works prior to making the PCB but this is the output I get:
... the mosfet tends to get warm so I think it starts to work in linear mode which is not adecuate . I really don't know what's wrong with it...
Ah, I've just taken a look at the MAX1674 but I can't see it can deliver 1A, it seems to be designed for a limit of 0.5A |
One problem with this type of converter, is that they can fail to start properly, and regulate if the full load is present when they switch on.
Maxim have an application note, about adding a MOSFET switch to the output of the device you are using, to disconnect the load, until the rail is stabilised (but the switch is only rated for 0.5A). The note is their number 254.
Also if your circuit is not carefully designed, you can get a worse problem where inputs to switching devices are 'undefined' during switch-on, till the processor starts, and begins driving the lines. Hence you may need to look carefully at adding very large value 'bias' resistors to ensure that every device is switched off, till the lines are being controlled.
These problems apply to all regulators of this type (not just the Maxim), and especially the possibility of silly loads during start-up, needs to be addressed before worrying about the regulator. Seperately, there is an 'issue' with trying to test such regulators on a plug-in board as you show. The instantaneous currents round the switcher, can be massive, and the resistance of such boards really makes them unsuited to testing above a very few hundred mA...
Best Wishes |
|
|
Hans Wedemeyer
Joined: 15 Sep 2003
Posts: 226
|
| You need a real circuit board to test DC-DC converters |
| Posted: Wed May 26, 2004 7:15 am |
|
|
| Mc wrote: | Hi!
Ah, I've just taken a look at the MAX1674 but I can't see it can deliver 1A, it seems to be designed for a limit of 0.5A |
No... not correct. Please read the data sheet.
MAX1674 is rated at 1 Amp.
MAX1675 is rated at 0.5Amp.
The MAX1674 start wil a load resistive load of 5 Ohm when set to 5V output.
The lowest guaranteed starting Voltage is 1.1 V However is you can afford a diode, MAXIM shows how that can help to start up even below 1.1V.
The really nice thing about this chip is " semi.'s are all in there" you add the inductor and filter cap. and the rest is in the chip.
One note: You CANNOT test DC-DC converters on that type of bread board, it will only give you BAD results, and worse it will mislead you into thinking DC-DC are not worth the effort... Do it right and it works very well.
I said you could make the DC-DC in the area of a nickel, well it can be done in less... here is one example I did recently. If I worked at the layout I think I could get two DC-DC in the area of a Dime. !
Sorry for the poor photo, I don't have time to make better today.
The complete DC-DC is outlined in red. |
|
|
Mc
Joined: 07 May 2004
Posts: 11
|
|
| Posted: Wed May 26, 2004 5:05 pm |
|
|
Thanks for all the advises!
I'll make the PCB and give it a try |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
