I am trying to do a project to make a DC to DC convertor to converte a Lipo that is 22.2v at 2100mah to output 12v at 1.5a.

I have some NTE966 regulators I wanted to use.

From the datasheet I am thinking I need something like this.
Datasheet can be found here http://www.nteinc.com/specs/900to999/pdf/nte966.pdf

I need a 0.33uF or larger tantalum or mylar capacitor on the input side and I am not sure how to get 10w resister.

Can someone give me some giudance here?

 

It is a little confusing whether you want a voltage regulator or current regulator.

For voltage regulation, you will be dissipating 15.3 W in the device (i.e., voltage drop X current). Look at the thermal specifications for that device. The junction to case resistance is 5°C per W. Thus, if the case were at 30°C, the junction would be at 76°C. The max. junction temp is 150°C. So far so good, right? HOWEVER, look at the junction to ambient resistance is 65°C, which means that without a heat sink, the junction temp would quickly go to toward 995°C (65 X 15.3). It will either thermally shut down or pop long before that occurs.

You will need a good heat sink to allow such a large power dissipation. Do you have the space for that? Your efficiency for the voltage conversion will be quite low. Can you use a different lipo configuration (e.g., 4S). Note: 22.2 V is the nominal voltage. When fully charged (e.g, 4.2 V/cell), the the voltage will be about 25V.

For current regulator, will your powered device handle 25 ? 10W resistors are available from Ohmite and other manufacturers at most parts dealers, such as DigiKey and Mouser. You will also need to consider the heat dissipation.

Have you considered a switching regulator?

John

 

You really need to look at building or buying a switching power supply.

A linear regulator will waste about half the power you have available, like John said.

Look at the attached schematic and simulation. It's a DC-DC "buck" converter that's relatively simple, yet 94% efficient.

L2 and C6 could be removed to make it even more simple, but at a cost of more noise on the output.

The parts used in the simulation are very specific.

No, you can't use the NTE parts to build it with.

 

Thanks everyone.

I will look at the one SgtWookie said. I just had the NTE parts allready and was think I could use them but I have no problem changing. I am learning more every day and this fourm is great. It has a lot of information.

I normally get my parts from Mouser and will look for the parts I do not have allready now.

Agian you guys are great. Thanks for the help.

Is there anything I need to now to build DC-DC "buck" converter.

 

Can you give part numbers for the L1 and L2 that you would use?

I was looking at
L1
http://www.cooperbussmann.com/pdf/4935b080-1881-43db-b73d-e2176f2c5e73.pdf
L2
http://www.delevan.com/seriesPDFs/4590.pdf

SgtWookie can you post the files from LTSpice here?

Or could I use a L4971 Datasheet here
http://www.st.com/stonline/products/literature/ds/5598.pdf

 

I'm attaching LT1171 12v buck supply.asc to this post.

You will need to have Linear Technology's LTSpice installed in order to view it.

LTSpice is a free download. Google is your friend.

Once you have the simulation loaded, click View => Bill of Materials => Show on Schematic.
Then View => Zoom to Fit

[eta]
Noticed that I didn't have a part number in one of the caps. Updated; here is the text BOM:

--- Bill of Materials ---

Ref.	Mfg.	Part No.	Description
C1	KEMET	C0603C105K8PAC	capacitor, 1µF, 10V
C2	--	--	capacitor, 100nF
C3	TDK	C575OX7RIH475M	capacitor, 4.7µF, 50V
C4	KEMET	T495X107K016AS	capacitor, 100µF, 16V
C5	Nichicon	UPL1J181MRH6	capacitor, 180µF, 63V
C6	Nichicon	UPL1V221MPH6	capacitor, 220µF, 35V
D1	--	MBRS360	diode
D2	--	MBRS360	diode
D3	--	1N4148	diode
L1	Coiltronics	CTX50-5-52	inductor, 50µH, 5A pk
L2	Wurth Elektronik	7447745039 WE-PD2 5820	inductor, 3.9µH, 2.6A pk
R1	--	--	resistor, 10.4K
R2	--	--	resistor, 1.2K
R3	--	--	resistor, 470
U1	Linear Technology	LT1171	integrated circuit

For the 100nF cap, just use a metal poly or ceramic.

[eta]
Instead of trying to find a 10.4k resistor for R1, use a 10k fixed and a 500 Ohm or 1k pot in series.

 

Thanks a lot.
Have the LTSpice allready loaded.

 

Keep in mind that you should not try to "breadboard" this circuit.

The internal oscillator of the LT1171 regulator runs at a pretty high frequency. You should make a proper circuit board. Keep the traces as short and as wide as you can.