Announcement

Collapse
No announcement yet.

Boost Regulator Modules and Circuits.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Boost Regulator Modules and Circuits.

    The Vallon VMH3CS, and no doubt other detectors, use boost regulators successfully to increase and regulate low battery voltages to higher voltages required by the detector circuit. I, for one, would like to use these readily available and low cost modules, together with high capacity Li-ion batteries to power any future PI detector designs.
    This looks like it could develop into a discussion and design topic outside of the Vallon thread; hence this new thread.

    Still on a learning curve, one question that I am pondering at the moment is whether there is any advantage (or disadvantage) in using a single high capacity pack of three) paralleled cells i.e. 3.7V at 7800mAh as against 11.1V 2600mAh series connected cells. The MT3608 converter output is adjusted to give 15V output with a detector load of 200mA. The published input voltage range is 2 - 28V

    It appeared to me that the conducted noise level, viewed on a scope at the output, was higher when running at 3.7V input. Would this be expected?

    One thing I learned yesterday is that series connected cells need battery protection circuits. I inadvertently left a detector switched on and while away the batteries died, or at least one of the two did. They were obviously not equally charged with one cell still reading 2V and the other 0V. The MT3608 will continue running to 2V with the other cell dead and SC. All attempts to wake up the dead cell have failed. The other is OK although running it down to 2V is 0.5V too low. Parallel cells equalise themselves presumably, so at the end voltage of battery life, all cell voltages are the same if you disconnect them. I plan on using protected batteries for forthcoming experiments and suitable candidates are Ansmann 3.7V 7800mAh or 11.1V 2600mAh battery packs. Both are three cells and available from RS Components. The interesting thing is that the 3.7V pack is very much cheaper, which takes me back to my question - which to use.

    Eric.

  • #2
    A few points:
    *Yes, 2 series cells or more always need a protection circuit, they can be easily bought online, or re-used from commercial packs, many of which are 2 or 3 cell, eg. cellphones, laptops...I have a number myself.
    *Step-up SwitchMode regulators work best when there's modest stepup, the currents are lower, losses in the switch, inductor, diode are lower, startup currents are lower, too. So I recommend stepping up 2 series Li cells rather than a single parallel pair, for example.
    *These modules can be optimised, datasheets are available, standard smpsu design rules/formulas etc apply, and you can optimise performance by tailoring them to your needs. Eg. if a module does 1A max output, but you're only ever going to use 200mA, then you could consider changing L value.

    Comment


    • #3
      Originally posted by Ferric Toes View Post
      It appeared to me that the conducted noise level, viewed on a scope at the output, was higher when running at 3.7V input. Would this be expected?
      According to the specificatioss of the SX1308 IC (the heart of the converter), efficiency at 3.7V would be around 85%, which is far from the 95% figure achievable with inputs above 5V. The frequency als drops to 0.8-0.9 MHz from the normal 1.1MHz. Both effects are sure to deteriorate the noise figure.
      Attached Files

      Comment


      • #4
        One of your first weapons when attempting to reduce output noise on the converter is a large value SMD ceramic capacitor as close as possible to the guts of the converter. By large I mean 4u7, 10u. And close means on the pads for whatever cap is already on the output, unless there's space to move it closer in. Major manufacturers produce smpsu IC app notes that contain great detailed layout information, showing what pcb tracks should be short, what can be longer. It depends on whether there is high-freq currents flowing, or just DC current, even if high A, for example.

        You will notice that there are a number of converter modules that have a fixed 5 Volt output. These should be suitable for applications where 4 x alkaline/zinc cells are intended, as any decent machine designed for a nominal 6 Volts should be fine at 5.0 V.

        For those of you contemplating custom battery/cell setups, you may not be aware that Li cells are available in AA size. But they're not normally called AA, to avoid mishaps, they are designated 14500 size. Capacities of the best ones are 800 - 850 mAh, quite a bit lower than you might expect based on the best 18650 cells. Their volume is 46% of an 18650, but capacity is 25%....

        Comment


        • #5
          ive been using a step up buck converter for some time now in one of my detectors and it works fine...i use several li-ion 3.7v un protected batteries but its always a worry if you leave detector switched on and batteries discharge too low...i guess its better to use protected type but more expensive....i used a large capacitor on output of conver ter and the detector is very stable.

          Comment


          • #6
            Replacing the plain 'bobbin' cored inductor with an equivalent fully shielded version can be another way to reduce interference generation.
            And if you're wanting good electrolytic caps for input / output filtering, Sanyo Oscon series (purple coloured) are excellent, and eat up noise impressively well for their size.

            For ease of manufacturing, standard inductors don't always have the maximum amount of copper wire wound on a former, and by rewinding with slightly heavier gauge, or bifilar/trifilar, it's possible to lower resistance, and increase efficiency a bit. Don't expect miracles, a lot of the loss is magnetic, in the ferrite.

            Comment


            • #7
              Dave

              Use a single cell protection pcb , not expensive.

              One example
              http://www.ebay.com/itm/Battery-Prot...GLvNL9ri4lGQow

              Comment


              • #8
                Hi Altra...at present im using a pack of Ni-Zn 1.6v AA batteries in my buck converter power supply cause of limited space....my boost voltage is around 16v with low noise..thanks for the link about battery protector.

                Comment


                • #9
                  ive just ordered some of those battery protectors from ebay.....thanks Altra.

                  Comment


                  • #10
                    Originally posted by Teleno View Post
                    According to the specificatioss of the SX1308 IC (the heart of the converter), efficiency at 3.7V would be around 85%, which is far from the 95% figure achievable with inputs above 5V. The frequency als drops to 0.8-0.9 MHz from the normal 1.1MHz. Both effects are sure to deteriorate the noise figure.
                    Thanks for that useful information. I have now ordered a protected 3 cell battery pack to give 11.1V to take the efficiency up to the 95% and push the operating frequency up. Efficiency seems to max at 200mA load as well, which is what I am drawing.

                    Eric.

                    Comment


                    • #11
                      Originally posted by Skippy View Post
                      One of your first weapons when attempting to reduce output noise on the converter is a large value SMD ceramic capacitor as close as possible to the guts of the converter. By large I mean 4u7, 10u. And close means on the pads for whatever cap is already on the output, unless there's space to move it closer in. Major manufacturers produce smpsu IC app notes that contain great detailed layout information, showing what pcb tracks should be short, what can be longer. It depends on whether there is high-freq currents flowing, or just DC current, even if high A, for example.
                      I measured the ceramic caps on the converter and both measure just under 4uF. I could piggy back a couple of 10uF ceramic caps on top if that would be an improvement. I guess it is a 'try it and see' job. The existing ones are mounted very close to the converter chip.

                      Eric.

                      Comment


                      • #12
                        Li-ion cells appear good in that they are 3.7V per cell, so you need less cells for a certain voltage, however, the best cells are expensive and have a capacity of around 2500mAh. I have just purchased a 3 cell 11.1V battery pack with protection which cost £37.82 from a reputable supplier. The cells are LG manufacture. Whatever the source, Li-ion cells are rather more touchy to use that other rechargeable types and need battery protection circuits.

                        I have just seen that Sanyo Eneloop Pro Nimh AA cells can be bought with a capacity of 2500mAh albeit with a cell voltage of 1.2, so I will need more of them, but, they are about £3 per cell.

                        Still using the MT3608 boost converter module to give me 15V, I can use 6 of the Nimh cells to give 7.2V which still gives top efficiency with a 200mA load and will do so until the battery drops to <5V, when the batteries will need recharging anyway. It looks to me that I would be better off with the latter, rather than the considerably more expensive and more critical Li-ion cells.

                        Eric.

                        Comment


                        • #13
                          Hello Eric,
                          I plan to run the AQ1B on those Sanyo Eneloop Pro NiMH. 8 cells will give me the required 9.6V and I can use a simple battery holder and charge them remotely on my smart charger. I don't mind the weight as I will need a bit of ballast in the enclosure.
                          I prefer the NiMH technology for their proven robustness plus the thought of a lithium setup (although unlikely) "toasting" my "significant investment" would be too much to bear......
                          Tony

                          Comment


                          • #14
                            I think the Sanyo Eneloop Pro are now Panasonic Eneloop Pro. I use them for my AKA Signum.

                            Comment


                            • #15
                              Originally posted by mschmahl View Post
                              I think the Sanyo Eneloop Pro are now Panasonic Eneloop Pro. I use them for my AKA Signum.

                              Yes, Panasonic.......I have just ordered some.

                              Comment

                              Working...
                              X