Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[PIC] Car Cooling System PSU Design Related

Status
Not open for further replies.

baileychic

Advanced Member level 3
Joined
Aug 2, 2017
Messages
728
Helped
56
Reputation
112
Reaction score
57
Trophy points
28
Activity points
7,033
[PIC] Car Cooling System PSU Design Related

I am doing a Car Cooling system using Peltier (4x) and PIC18F46K22. 2x LM35 are used to sense room temperature and Peltier temperatures. The coding part is almost complete and working fine in Simulation.

For Peltiers connected in Parallel they need a max 15V@15A supply.

I have a need for 12V 3A supply for cooling fans and a 5V 1A supply for MUC+LCD.

There is no time to design a DC-DC Converter and a Battery Charger at this time and hence I am going to use this DC-DC Converter which gives 24V DC from 24V DC input.

https://www.ebay.com/itm/WaterProof...-DOWN-Power-Converter-Regulator-/152975775363

I have attached a circuit which I have designed. It uses a 24V 270W Solar Panel to feed the DC-DC Converter to get 15V DC@20A.

I have used a logic level mosfet and diodes to drop this 15V DC to 14.7V to charge the Series connected 12V 20Ah batteries in Parallel.

I have not yet finalized the used diodes or logic level mosfet. I might change them to better ones with low Vf and low Rds(On) to minimize power loss.

The batteries if 20Ah type needs say 0.1 * 20A = 2A charging current.

My questions are is the circuit in the attached PDF good for a quick solution?

I have no issues with the 12V and 5V DC supplies part in the Power Supply.

For batteries B1 and B2 I have to apply not more than 14.7V boost charging voltage and 11.3V trickle charge voltage. I am planning to use PWM to control the charging voltage and or current.

So, can I use the circuit in PDF?

I have questions related to D1, D4, D5 diodes.

D1 has to block the 24V of battery from getting into Solar Panel.
D4 and D5 should help in charging Battery B2.
The D3 and D4 diodes are for blocking reverse flow of voltage/current from Batteries to DC-DC Converter output.

Will D5 diode gets forward biased while charging the B2 battery and affect the B1 battery charging?

If there are any issues then how they can be resolved?
 

Attachments

  • PSU2.2.PDF
    38.5 KB · Views: 80

Hi,

I didn't go through the complete circuit.

A couple of issues.
You say this circuit is for "cooling". But then there are a lot of diodes in the power path...and additionally a FET to drop voltage (it seems to be unregulated, so I ask myself how it can work).
Each voltage drop in a power path causes power dissipation = heat according P = V x I.
Maybe you need another cooling circuit for your cooling circuit.

B1 has a diode in series, thus it can only be discharged, but never charged...
The whole expected function is a riddle to me.

Q1 will never be able to output 14.7V or even close to it with a MCU input signal. With N-Ch Mosfets V_gate needs to be higher than V_source.

Zeners as protection in analog measurement signals are not recommended. The protection level is doubtful and they cause (huge) measurement errors.

1000uF at the output of a 7805 is useless (although frequently seen in amateur circuits).
* Usually 1000uF capacitors are not good for high frequencies
* For low frequencies they make no sense here, because the voltage regulator will be faster
* and for energy storage they make way more sense at the regulator input.

Klaus
 
Doesn't look at all right to me.
I'm not sure what any of the circuit really does. D2 seems to have no purpose at all. If the DC-DC converter is to step down the PV voltage so it can charge the batteries individually, the current steering is wrong and B1 will never receive any charge. The MCU voltage has to go above about 19V before the MOSFET turns on and the 15V sense signal will always be high because it is fed from both PV and battery sources. Does the fan voltage (12V) really need regulating at all and does the MCU really need a 1A supply?

An MPPT device with two isolated outputs to charge the batteries in series would be more practical and efficient.

Brian.
 
Corrected in previous post

There is no time to design a DC-DC Converter and a Battery Charger at this time and hence I am going to use this DC-DC Converter which gives 24V DC from 24V DC input.

There is no time to design a DC-DC Converter and a Battery Charger at this time and hence I am going to use this DC-DC Converter which gives 15V DC from 24V DC input.

I don't want any frequency in DC output.

DC freq = 0

15V DC I get from DC-DC Converter. I can add a 4500uF 50V at the output of DC-DC Converter to smooth it.

Issue are with D1, D4 and D5.

Why can't I get 14.7V after MOSFETs and diodes? Because of Mosfet and diode drops? Are the diodes D3 and D4 really required for reverse Voltage/Current blocking? Can't the Mosfet do that?

If I remove D5 then Batteries will charge. Can I also omit D3 and D4 (I don't want the 24V from batteries flowing into the DC-DC Converter output).

Regarding driving the MOSFET, I will always have 24V at the battery output whether there is Solar Panel power or not that is Batteries will be kept fully charged and so I can make a small NPN transistor circuit which is driven by MCU logic level and the NPN will, in turn, drive the MOSFET gate.

Regarding heating issues what alternatives can I use to avoid heating? I am not worried about Power loss.

Is there a better and simpler alternative to get 15V@20A and also charge the 2x 12V batteries without heating issues?

I will quickly make a new circuit with NPN transistor which drives the Mosfet.
 

Why can't I get 14.7V after MOSFETs and diodes? Because of Mosfet and diode drops? Are the diodes D3 and D4 really required for reverse Voltage/Current blocking? Can't the Mosfet do that?
There is a clue in the MOSFET symbol about that!

I can't understand why you need the 15V. If it is to charge the batteries you can't do it while they are in series, you need >24V to do that.
If you do use a block DC-DC converter rather than a step down regulator, it will almost certainly use some kind of switching circuit with a diode in series with its output already so you shouldn't have to worry about back-flow.

Brian.
 
I fixed the MOSFET circuit with High Side Driver but still, it doesn't work as needed.

Find the new circuit which is attached.

In simulation the HO signal drops gradually to less than 5V. What is the issue?

In the simulation the Battery models that I have used are not re-chargeable type. They provide a constant 12V (12V + 12V = 24V) pressure.

- - - Updated - - -

Okay, lets assume that the DC-DC Converter has a diode already and if I remove the diodes in the Battery charging circuit then how can I charge series-connected Batteries in Parallel?

15V @ 15A is needed for the 4x Peltiers and that is why we are using the step-down DC-DC Converter.

Any alternate and cheaper solution?

Should I work out a few DC-DC Converter euuations and make a new DC-DC Converter with using fixed PWM which then gives say 24V from say 36V Solar Panel?


Alternatively can I directly charge the Batteries (24V, 12V x 2) from the Solar Panel by using a Mosfet with PWM and control the charging voltage/current? Then the DC-DC Converter will not be used for charging the Batteries. When Solar Panel Power is sufficient then the batteries will get charged and the DC-DC Converter will give 15V from Solar Panel and also 24V of Batteries.
 

Attachments

  • PSU2.4.PDF
    43.1 KB · Views: 59
Last edited:

Hi,

Most probably the high side driver does not work as expected .... because it can only work with duty cycle less than 100%, because of it's bootstrap circuit. It is explaind many times here in the forum and in any high side driver application note.

I see no urgent need to externally smooth the output of a DC- DC converter. As the name says: it already should be DC. It's not a PWM output.

Batteries don't provide constant output voltage. For a 12V Pb battery you should calculate with 11V ... 14.4V.

how can I charge series-connected Batteries in Parallel?
Series contradicts parallel --> thus I say it is impossible. Either the batteries are in series or they are in parallel.

15V @ 15A is needed for the 4x Peltiers and that is why we are using the step-down DC-DC Converter.
Did you read peltier application notes.
I'd rather use constant_current than constant_voltage.

Should I work out a few DC-DC Converter euuations and make a new DC-DC Converter with using fixed PWM which then gives say 24V from say 36V Solar Panel?
When there is no time to design a charger, then why not but it? There are cheap charger for almost any combination of solar cells and batteries.

Klaus
 

Does your application permit charging one battery at a time?
With some MCU intervention and some current steering it is possible to charge one at a time but you will not be able to use the '24v' while they are in charging mode. If the batteries are to provide power when there is no PV available it should be possible to use PV to get 24V and switch to batteries as the PV voltage drops.

Brian.
 
Hi,

I'd go for
* either 12V system
* or 24V system.
don`t go the mixed way.

I tend to a 24V system because of the lower currents = lower expectable loss = less heating.

Klaus
 

Simple method to charge two batteries in series, from one power supply. Drive a half-bridge so it alternately provides current to each battery.

charge 2 bat in series via NPN PNP 2 diodes 14V supply 555 IC pulse generator.png

The diodes are necessary.

One battery can be absent, while the other battery charges.

Select bias resistor values so transistors turn on sufficiently.
 

@BradTheRad

Thank you for the circuit. Can I drive the Complementary transistors in your circuit with PWM from MCU (PIC) instead of using the NE555 and use variable duty to get trickle/boost charge voltages? I will feed 15V@20A output of DC-DC Converter to the Complementary transistors to get 14.7V boost charge voltage for the Batteries.
Instead of using BJTs, I will use Very low Rds(on) MOSFETS. Is that okay? Because using MOSFETS, I don't have to worry about Biasing resistor values or base currents because MOSFETs are voltage operated devices and I can easily use a 33 Ohms resistors for the Gate Resistors.

I will use a > 100A MOSFETs and the Batteries should not draw more than 2A. So, should I set the PWM duty according to that? 20Ah battery's nominal charging current is 0.1 * 20A = 2A. I don't want to allow > 2A to flow into battery and make it explode.

- - - Updated - - -

Can I use this diode to replace all Power diodes in my circuit?

https://www.mouser.com/datasheet/2/240/DPG30P400PJ-1546700.pdf

Check the above datasheet page 1. I will just use Pins 1 and 2. My current will not exceed 10A and so I will use single diode.

Find the new circuit which I made based on the BradTheRad's circuit. Is it okay now? Check the graph and reply.


Is this PCB track width okay?

https://www.7pcb.com/trace-width-calculator.php

I will use only 2 layer board that is top and bottom layers and there will be via stitching on all Power lines. I will use 400mils track width. The above calculator (image) shows 8.4 mm (331 mils).
 

Attachments

  • PSU-3.0.PDF
    54.2 KB · Views: 61
  • PCB Track Width-1.png
    PCB Track Width-1.png
    135.3 KB · Views: 82
Last edited:

I don't want to allow > 2A to flow into battery and make it explode.

It's okay to use mosfets with pwm although their low On-resistance may pass several (perhaps too much) Amperes to the batteries. More so when batteries are at low voltage. Therefore it's a good idea to limit current (in addition to adjusting pwm duty cycle).

My schematic illustrates a concept but it does not account for all situations. Example, what happens when the control device is off? Does a transistor turn on? Does it expose a battery to full supply voltage?
Etc.
 

How to limit the charging current to the battery? I don't want it to be more than 2A.
 

How to limit the charging current to the battery? I don't want it to be more than 2A.

I think it's okay if you apply 4A at 50 percent duty cycle to each battery. (Don't know about 8A at 25% duty. Or 16A at 12.5% duty.)

It's efficient if you maintain the power source at just a volt or two greater than battery voltage. It's not necessarily easy to do since a discharged battery can be 10 or 11V, then it increases to 14.5V as you charge it.

There is nothing wrong with using PWM. To limit pulses to 2 or 3 or 4 A, add some filtering. It can be choke and/or capacitive. The choke is along the lines of a buck converter. It's reasonably efficient.

There is no need to eliminate ripple. In fact if you add too much choke, it tends to average pulse amplitude, and the voltage becomes too low to charge the battery.
 
Now there is an additional requirement and that is I have to control the load power by using a MOSFET switch. The load power is supplied by the +15V output of the DC-DC Converter. So, the Load Power Control MOSFET switch needs to be in series with it but that will drop some 0.4 to 0.6V I guess and that will reduce the voltage to Peltiers rated at 15V.

So, how can I add this requirement with getting 15V for Peltiers?
 

Hi,

If you want to regulate the power of a peltier you need to control the voltage...somehow
You can't keep voltage at constant 15.0V.

Again my recommendation:
* Try to focus on current instead of voltage.
* For best efficiency use DC instead of pulses.

Klaus
 
I found that Peltier can run from 14.5V also so the voltage drop is not an issue now. The only issue now is how can I use a MOSFET high-side to control the load power that is +15V output from the DC-DC Converter which provides all the power to the device?

Please see the new attached circuit. Can I use the Q1, Q2 circuit itself to get the 15V or 14.5V main DC-DC Converter power to the device (Peltier and 12V, 5V regulators) by adding a capacitor across "BAT CHG" connector and GND? The "BAT CHG" connector is already giving 14.5V output and so I can smooth it with a say 100uF 50V and take it as the input for Peltier/LM2596? That is by keeping only Q1 On when battery is not charging and switching Q1/Q2 complementary while Battery is charging.
 

Attachments

  • ECCS-PSU.PDF
    52.7 KB · Views: 80
Last edited:

Hi,

Why not simply switching low side = GND of the peltier?
Why not using a dedicated Highside_switch?
Why not using standard P_channel circuits to control high side?

Klaus
 
How to control PNP at high side (supply +15V) via NPN at low side which receives 5V bias.

control PNP high side 15V supply via NPN low side bias 5V.png

On the other hand, can you install the load at the collector leg of an NPN? Or does your arrangement only permit the Peltier modules to be grounded to the 0V supply rail?
 

Okay, I will add an N-Channel Low-Side MOSFET between Peltier's -Ve pin and GND connection. That solves my issue.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top