EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: andersendr on June 15, 2010, 12:41:43 am
-
Hello EEVblog Forum, I am wondering if this sort of thing is even possible, I am using a Lithpoly battery at 3.7 volts and would like to get a +-25 volt supply to run an op-amp to go from 0v to 5v up to 16v to 25v. I have the op-amp design down but my question is if I can generate a +-25v supply from a 3.7v lithpoly?
This is sort of a retrofit circuit to an existing circuit and I would like to keep the existing circuit. The existing circuit connects right to the 24v battery system and shows battery voltage. I am trying to replace the 24v connection by recreating it using a PWM signal and the op-amp configuration. Or would it be easier to just make 5LED's in the same form factor and just control each LED as ON or OFF. The part I would loose is the auto dimming of each LED if I had to replace them with an ON/OFF configuration.
Here is the current circuit, I am trying to eliminate the B+ and B- and replace them with the op-amp connections.
(http://dellslivesteamers.org/BatteryVoltage.png)
Here is the direct link for those who would like it.
http://dellslivesteamers.org/BatteryVoltage.png
-
A few blogs ago, Dave covered a type of switched mode power supply called the 'buck' regulator. There is also a design called the 'boost' regulator. As the name implies, it boosts the voltage. There are quite a few of them around, this is an example from Linear, but many other manufacturers make them too (Maxim, TI, Analog etc)
http://en.wikipedia.org/wiki/Boost_regulator
http://www.linear.com/pc/productDetail.jsp?navId=H0,C1,C1003,C1042,C1033,P91188
If it were low current, and not vulnerable to noise, you could use a simple 555 oscillator and Cockroft-Walton multiplier: http://en.wikipedia.org/wiki/Voltage_multiplier
-
What you have is an LED voltmeter connected from B+ and B-. It was originally designed to work off 24V.
You need a DC voltage multiplier or x6 or more, to replace B+.
This should provide the leads to make this circuit work:
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1017
Hello EEVblog Forum, I am wondering if this sort of thing is even possible, I am using a Lithpoly battery at 3.7 volts and would like to get a +-25 volt supply to run an op-amp to go from 0v to 5v up to 16v to 25v. I have the op-amp design down but my question is if I can generate a +-25v supply from a 3.7v lithpoly?
This is sort of a retrofit circuit to an existing circuit and I would like to keep the existing circuit. The existing circuit connects right to the 24v battery system and shows battery voltage. I am trying to replace the 24v connection by recreating it using a PWM signal and the op-amp configuration. Or would it be easier to just make 5LED's in the same form factor and just control each LED as ON or OFF. The part I would loose is the auto dimming of each LED if I had to replace them with an ON/OFF configuration.
Here is the current circuit, I am trying to eliminate the B+ and B- and replace them with the op-amp connections.
(http://dellslivesteamers.org/BatteryVoltage.png)
Here is the direct link for those who would like it.
http://dellslivesteamers.org/BatteryVoltage.png
-
Using a 555 with a cockroft-walton multiplier is not a bad idea. Just watch your current draw. The easier solution, however, would be to make a simple boost switcher. Use a 34063a as the switcher chip. Get the data sheet and there are simple circuits to design the converter http://www.onsemi.com/pub_link/Collateral/MC34063A-D.PDF
The chip is cheap, and the whole thing should be under $5.
paul
-
Consider another type of bar graph design, here's one more independent of input voltage. In the othe design, even if B+ is rated 24 Vdc or so, a zener diode there limit the operating voltage to at least 9V, not sure why its there as it seems wasteful.
Here's a similar one that will dim LEDs before engaging the next one, and is independent of the driving voltage although it will require at least 3V to drive the LEDs and Vbe drop:
http://electronicdesign.com/article/components/recursive-circuit-for-bar-graph-led-displays-and-s.aspx
(http://archive.electronicdesign.com/files/29/6246/figure_01.gif)
-
This circuit is ... useless.
If you use step up transformer output will always be 24V before battery goes dead.
I presume this is a low capacity battery under 1Ah.
If voltage meter is not critical I would omit it and stick with protective battery circuit with under and overvoltage cutout.
-
The circuit was designed to be directly connected the battery. I am just trying to fake out the batter with a op-amp but need 0-24v range to do so. I have decided just to run the 24v from the battery up to the controller so this can still work then.
-
Hi,
I'm still very much a noob but I think I can explain why Pyr0Beast says "This circuit is ... useless."
I assume you want the circuit you want to save is a battery state of charge indicator that is part of the instrument panel of some piece of mobile machinery
and that this circuit has 5 LED that each fade out as state of charge decreases
for some reason you will be powering the machine (or just the indicator) off a 3.7V battery
using a boost type smps won't work because the voltage output of it will not have the same values as a lead acid 24V battery
from what I've read off wikipedia, typical values for LiPo cells are 3.0V minimum to 4.23V maximum
while a 24V lead acid battery (doubling the values I found for a 12V battery) are 21V minimum to 25.6V maximum
that is open circuit voltage, if the battery is loaded I assume the values would change, also the state of charge value is not changing linearly with the voltage
see this graph http://en.wikipedia.org/wiki/File:Lead-acid_voltage_vs_SOC.PNG
I guess that LiPo cells are also non-linear and that curve is not the same as lead-acid batteries so that makes direct voltage boost inaccurate
also I think regulated boost smps will always output 24V until the input voltage drops to low at which point the output will sharply fall to zero
I think to get any kind of accuracy and not spend more than the value of your entire machine on SOC indicator R&D I would suggest the following approach
get a 12$ arduino, hook one of the ADC to the voltage battery, add appropriate input protection and noise filtering (this is the part where I am a noob and don't know what to put there !)
remap the value of the ADC to have 4.2V = 100% 3.0V = 0%
or even better find the characteristic state-of-charge curve of the LiPo cell you are using and map that to the ADC values
and for output either
1 generate 24V "somehow" (maybe that not so useless now boost circuit) and modulate it according to the actual recorded SOC value this will give you analog-looking indicator values and you won't have to modify the original SOC circuit
2 override the SOC circuit and directly drive the LEDs off digital I/O pins this will give you 5 discreet state of charge indicator values
3 override the SOC circuit and directly drive the LEDs using 5 of the 6 PWM output pins this will give you analog-looking indicator values and is the simplest cheapest design I can think of right now
-
The circuit is not a boost converter which is required to increase 3.7V to 25V.
This isn't the best way of going about it anyway, if you want to power a load of LEDs from a 3.7V battery, connect them in series.If they're white, the battery voltage is too low so you'll need a boost converter anyway, but unless you're using a transformer, it's more efficient to increase the voltage less and connect the LEDs in parallel.