Author Topic: Overvoltage problem from Sony 7.2V Li-ion battery fresh off the charger.  (Read 10744 times)

0 Members and 1 Guest are viewing this topic.

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
Is it possible to protect a device from the peak voltage of a freshly charged 7.2V Sony Camera battery which is at 8.4V to keep the voltage at the 7.2V rating.

I would like a small portable simple pcb for the battery as this is for powering a hand held metal detector with a max allowed voltage of 7.2V.

The 8.4V is stopping me from using this battery system and need help with circuit ideas.
Thanks for any help!
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
You can get very low dropout regulators like this:

http://www.st.com/web/en/resource/technical/document/datasheet/DM00063302.pdf

They'll cap the voltage wherever you choose. It has a dropout of 125mV (typical; 300mV max), so if the battery voltage drops below 7.2V, this device will output a voltage 0.125V less than its input, which may or may not be acceptable in your case. One-part solution :-)
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
Is it possible to protect a device from the peak voltage of a freshly charged 7.2V Sony Camera battery which is at 8.4V to keep the voltage at the 7.2V rating.

Your title is slightly misleading as the 8.4 V from the battery is not over at all, it is the normal and expected voltage for this and all such lithium ion batteries. If a device is designed for a lithium ion battery and it sees a voltage of 7.2 V it will think the battery is flat and will give a low battery warning.

But perhaps your device was originally designed for some other battery system like alkaline cells and this is why you are worried?
« Last Edit: October 06, 2014, 01:13:50 am by IanB »
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
You are correct the 8.4 V from the battery is not over for the battery but it is over for my detector as it has a max input of 7.2V.

The detector is a Minelab SD2200 and was designed to use a 6V SLA so the 7.2 is a boosted voltage to help smooth the sound and better depth.

I did forget to say the detector requires 850MA to run and needs 1000MA at start up.

I have been using a Castle BEC switching regulator set at 6.8V but it is causing EMI with my detector. http://www.castlecreations.com/support/documents/cc_bec_user_guide.pdf

Coiltek makes a regulator but is very pricey at 165.00 so I would like to build my own.

I have looked at Zener diodes but not sure if that is the way to go so looking for help here.
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
I believe you can use LDO regulators in parallel and they'll share current. So a couple of the parts I recommended earlier would still work. But it's probably better to use an LDO with an external transistor drive, like the MIC5159. That way, the dropout is as low as you want, you choose the transistor.

You may wish to look at the parts inside and see if it's really a maximum of 7.2 -- if I design a device to work off a 12V supply, I'm generally going to use parts rated for 15V, for example. You may not need to worry.
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1380
I believe you can use LDO regulators in parallel and they'll share current. So a couple of the parts I recommended earlier would still work. But it's probably better to use an LDO with an external transistor drive, like the MIC5159. That way, the dropout is as low as you want, you choose the transistor.

You may wish to look at the parts inside and see if it's really a maximum of 7.2 -- if I design a device to work off a 12V supply, I'm generally going to use parts rated for 15V, for example. You may not need to worry.
Paralleling LDO's can be very touch and go.  I paralleled 2 linear regulators (cant remember if they were 7805's or something else) with a series 0.1 ohm resistor for each to help them load share.  It worked fine at first, but after a while it ended up conditionally stable, oscillating at ~30 MHz when the load wasnt heavy enough.  Made for all sorts of strange behavior with a bunch of 7400 series logic.  Couldnt find the problem in the field with a fluke scopemeter with only 2 (or 10? cant remember) MHz of bandwidth.  Still took a few days in the lab to figure out the true source.
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
Thank you all for the feedback and good ideas.

rs20 I would look inside but mostly white paint in there.

The MIC5159 is interesting but the  parallel is the part I have to wonder how you get them to share equally?
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
Thank you all for the feedback and good ideas.

rs20 I would look inside but mostly white paint in there.

The MIC5159 is interesting but the  parallel is the part I have to wonder how you get them to share equally?

Forget paralleling, I shouldn't have suggested it. In any case, with the MIC5159, you don't need to parallel anything. One MIC5159 + a suitable FET can do 10A if you want.
« Last Edit: October 07, 2014, 02:51:16 am by rs20 »
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
rc20 I found this LT3080 on YouTube under LDO which looks promising also.

http://pdf1.alldatasheet.com/datasheet-pdf/view/349922/LINER/LT3080.html

Your idea of the MIC5159 + a suitable FET is a good one as I was not thinking FET.

I think I got brain locked with Zener Diodes so as to not waste power.
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
rc20 I found this LT3080 on YouTube under LDO which looks promising also.

http://pdf1.alldatasheet.com/datasheet-pdf/view/349922/LINER/LT3080.html

Fairly poor choice because the LT3080 requires a supply input at least 1.35V above the output voltage ("VCONTROL Dropout Voltage"). (You only get 350mV dropout with multiple supplies. Your own FET with a MIC5159 can get 100mV drop easy.) So even when your battery is at 8.4, your output will be less than 7.2V. When the battery is at 7.2V, your output will be under 6V. That's 15% of your power gone. The dropout of the MIC5159 is as low as you want it to be, you choose the FET.

Your idea of the MIC5159 + a suitable FET is a good one as I was not thinking FET.

I think I got brain locked with Zener Diodes so as to not waste power.

I don't understand what you're saying here -- Zener diodes are only part of a solution, and not a particularly non-power-wasting solution at that; feel free to recreate the MIC5159 from individual discrete parts if you want, but I doubt you'll be able to achieve the same performance or efficiency without a lot of effort. Your device is consuming 8W, it's not worth even thinking about efficiency really... anything vaguely sane will consume orders of magnitude less power than your device.
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
rs20
I had read somewhere long ago that Coiltek company was using a diode to regulate this same battery setup for my machine that was switchable between 6.7V and 7.2V but that unit was priced at around 400.00 so I did not purchase it and could not find any info on the build. Thats the long story of the diode that I do not have any info on just what I read.

http://arizonagoldprospectors.com/cool-tools-lion.htm

I need to spend more time reading up on your MIC5159 idea but I am slow at reading data sheets as I am new at building from scratch.

I have a lot of personal time around PacMan PCB repair and an auto mechanic but data sheets are fairly new to me.
Thank you so much for your time and help!
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
rs20
The MIC5159 is an obsolete item in the US I can not find one!

I got the idea so will look on Digi-Key for something I can try like the NCP565D2TR4G.
Thanks

 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19345
  • Country: gb
  • 0999
You are correct the 8.4 V from the battery is not over for the battery but it is over for my detector as it has a max input of 7.2V.

The detector is a Minelab SD2200 and was designed to use a 6V SLA so the 7.2 is a boosted voltage to help smooth the sound and better depth.

I did forget to say the detector requires 850MA to run and needs 1000MA at start up.

I have been using a Castle BEC switching regulator set at 6.8V but it is causing EMI with my detector. http://www.castlecreations.com/support/documents/cc_bec_user_guide.pdf
Try adding a filter to the output of switching regulator.

 

Offline bills

  • Frequent Contributor
  • **
  • Posts: 350
  • Country: us
or you could do it simple, put 3 1n4007 diodes in series and you will have a 1.8 volt drop.
Each diode will give you a .6 volt drop.
« Last Edit: October 14, 2014, 08:22:21 am by bills »
Never argue with idiots. They drag you down to their level and beat you with experience.
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
or you could do it simple, put 3 1n4007 diodes in series and you will have a 1.8 volt drop.
Each diode will give you a .6 volt drop.

With the obvious disadvantage that once the battery gives 10% of its charge, the battery voltage will drop to the nominal 7.2, the voltage after the diodes will be 5.4, and it'll probably cut out due to "low battery".
 

Offline sleemanj

  • Super Contributor
  • ***
  • Posts: 3020
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
What you want, is a small buck converter.

Like these ones I sell in NZ, but I'm sure you can find on ebay etc:
http://sparks.gogo.co.nz/Miniature-DC-DC-Step-Down-Buck-Converter-248.html



Max input 23v, output 1 to 17v, single turn trimmer to set the output voltage, 1.8A or so continuous.  17x11mm, 1.4g; MP2307DN based.

Note that this doesn't protect your lithium batteries from over-discharge.
~~~
EEVBlog Members - get yourself 10% discount off all my electronic components for sale just use the Buy Direct links and use Coupon Code "eevblog" during checkout.  Shipping from New Zealand, international orders welcome :-)
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
Guys thank you for your input, what do you think about this little circuit?
From: http://www.radio-electronics.com/info/data/semicond/zener-voltage-reference-diodes/applications-circuits.php
A Zener diode circuit that uses a series pass transistor. A simple circuit is shown below and here the transistor is used as an emitter follower.
 
Zener diode circuit for a simple regulated power supply
When utilizing this circuit, the current required from the Zener resistor potential diver should be calculated. This is the emitter current from the transistor divided by the gain.
When choosing the Zener diode voltage, it should be remembered that the emitter voltage will be lower than the Zener voltage by the amount of the base-emitter voltage - 0.6 volts for a silicon transistor.
To be honest I am having trouble wrapping my head around a Zener diode and if they will drain power after the battery drops at or below the Zener voltage and it is no longer allowing current to flow in reverse to ground.

If my thinking is correct then this little circuit should work to drain off the excess voltage and stop draining when voltage is at or below Zener voltage. Am I correct or out in left field?

Sleemanj it is my understanding that the buck converter creates a lot of emi when switching off, is this true about the emi as that is the problem I have now with the bec.
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
Guys thank you for your input, what do you think about this little circuit?
From: http://www.radio-electronics.com/info/data/semicond/zener-voltage-reference-diodes/applications-circuits.php
A Zener diode circuit that uses a series pass transistor. A simple circuit is shown below and here the transistor is used as an emitter follower.
 
Zener diode circuit for a simple regulated power supply
When utilizing this circuit, the current required from the Zener resistor potential diver should be calculated. This is the emitter current from the transistor divided by the gain.
When choosing the Zener diode voltage, it should be remembered that the emitter voltage will be lower than the Zener voltage by the amount of the base-emitter voltage - 0.6 volts for a silicon transistor.
To be honest I am having trouble wrapping my head around a Zener diode and if they will drain power after the battery drops at or below the Zener voltage and it is no longer allowing current to flow in reverse to ground.

Best case dropout you can get with that circuit is 0.7V. Extremely low dropout is a necessity for your application. I gave you the low-drop out solution with a chip driving an external PFET many messages ago. Low EMI, low dropout, low quiescent current, two cheap parts.
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
rs20

The MIC5159 you talked about I can not find as it is obsolete? Also how would I handle Input voltage range: VIN 1.65V to 5.5V where I have 7.2V?

Is there another you talked about?
Thanks
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
Thanks for all the help!

I am going to try my luck with the MIC29302WT TO-220 adjustable version.

Regards to all,
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
rs20

The MIC5159 you talked about I can not find as it is obsolete? Also how would I handle Input voltage range: VIN 1.65V to 5.5V where I have 7.2V?

Is there another you talked about?
Thanks

"But it's probably better to use an LDO with an external transistor drive, like the MIC5159" is what I said; it's not like the MIC5159 is a unique part. There are probably thousands of such parts. In any case, the one-part solution you've selected has a 250mV dropout (typ) at 1.5A, which is OK. You can calculate out how much sooner the metal detector will turn off due to "low battery" due to this dropout, and figure out whether going for a two part solution (with something like the MIC5159 but with suitable voltage range and not obsolete) is worth the extra battery life for you.
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
Rs20,
I thank you for all your help and I understand the MIC5159 may be a better choice however from reading the datasheet I feel it is way above my limited ability to design and complete a working circuit.

The only MIC5159 I find is a surface mount from Digi-Key and I am not able to work with that mount. I will keep looking and trying to master this part and data sheet.

I am pleased that you say my MIC29302WT selection is “OK” which to me is a good thing.

I do not feel my detector will turn off as it was designed for a 6V SLA battery and the Li-Ion 7.2 would over discharge long before the detector would stop working.

I am not good with math just my hands as I can build a 300mph race car but can’t figure the angles of a square.

Thank you as I really appreciate all your help and hand slapping!

Regards,
Sprig
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
Rs20,
I thank you for all your help and I understand the MIC5159 may be a better choice however from reading the datasheet I feel it is way above my limited ability to design and complete a working circuit.

The only MIC5159 I find is a surface mount from Digi-Key and I am not able to work with that mount. I will keep looking and trying to master this part and data sheet.

To be clear, I agree with you that the MIC5159 is a poor choice, especially since its supply voltage does not go as high as 8.4V. What I meant was something like the MIC5159. Anything under this entire heading on digikey: PMIC - Voltage Regulators - Linear Transistor Driver. They're only very slightly more difficult to use than standalone regulators, you just combine it with a separate transistor and the two parts together act as an LDO regulator. How do you combine them? Just look at the example schematic in the datasheets, it's right there. The reason performance is so much better is that manufacturing processes that make good ICs and manufacturing processes that make good power FETs are very very different; by buying two separate chips made on separate processes, you get a much better performing part for demanding applications.

Regarding my comment that it's OK, don't forget my followup comment that battery life will be worse with the MIC29302WT than with a better two-part solution. It would be wise to calculate just  how much worse it will be so that you can make an informed decision as to whether your happy with the compromise. I haven't done this calculation, so don't trust me too much when I say it's "OK"!
 

Offline Mr 12Topic starter

  • Contributor
  • Posts: 19
Rs20
I think I found an acorn?

The link you provided lead me to MIC5156BN after I drilled down a bit looking for something that I could work with and would take my supply voltage.

There is a nice diagram of the adjustable configuration on the datasheet.

http://www.micrel.com/_PDF/mic5156.pdf

Now I just need to figure out the math equation for the resistors I need to get 7.2V output?
Regards
 
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
Rs20
I think I found an acorn?

The link you provided lead me to MIC5156BN after I drilled down a bit looking for something that I could work with and would take my supply voltage.

There is a nice diagram of the adjustable configuration on the datasheet.

http://www.micrel.com/_PDF/mic5156.pdf

Now I just need to figure out the math equation for the resistors I need to get 7.2V output?
Regards

Reading the datasheet,

"The MIC5156 requires an external gate drive supply to provide the higher voltage needed to drive the gate of the external MOSFET. The MIC5157 and MIC5158 each have an internal charge pump tripler to produce the gate drive voltage."

This is because the MIC515x series uses N-type transistors, which require gate voltages higher than supply. If you use P-type transistors, you won't have this issue. So either find a part that uses P-type transistors, or make use of the MIC5157 or MIC5158 instead of the MIC5156.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf