Maker Pro
Maker Pro

Eye Movement Desensitization Rehabilitation device

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
The dual op-amps I've suggested are 8-pin devices. You can use one op-amp as the Schmitt trigger for the oscillator and the other as the inverter to drive the second transistor.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
Price is not a problem unless it costs something like £100.

The OPA2350 has a wider output voltage and current range is there anything wrong with that?
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
The OPA2350 looks good. It has a much lower input offset voltage, which isn't really important, and it's much faster, which also isn't important. It draws a lot more supply current, but 15 mA isn't a big deal compared to the current the motors could be drawing. The output swing is good.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
I probably should have mentioned this sooner, it will be powered by 3x AA batteries will i need any bypass caps?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Yes. Put a 0.1 µF ceramic across pins 4 and 8 of the dual op-amp as close to the IC as possible, and 10~100 µF electrolytic across the rails somewhere else.

The emitter follower will drop a minimum of about about 0.7~0.8V so you won't get quite 4.0V from a 4.5V supply.

Because of the voltage drop in the emitter follower, and the minimum voltage required before the motors will even start, you will have a non-functioning deadband at the minimum end of the voltage adjustment potentiometer. This can be removed if you insert a resistor (probably around 15% of the potentiometer's resistance) between the "minimum" end of the potentiometer and VCC.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
I think I need to find a germanium BJT apparently they have a very low Vbe.
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Well, that's an interesting idea! Yes, germanium transistors do have lower Vbe than silicon transistors. They aren't very common nowadays, but NTE (http://www.nteinc.com) make them. You can also get them through eBay, but buying electronic components through eBay is a bit like Forrest Gump's "box of chocolates" - you never know what you'll get.

Here are the suitable germanium PNPs from NTE (http://www.nteinc.com/Web_pgs/Germanium.html):

NTE102A (ok) http://www.nteinc.com/specs/100to199/pdf/nte102a.pdf

NTE158 (good) http://www.nteinc.com/specs/100to199/pdf/nte158.pdf available from Radio Shack for USD 5.09: http://www.radioshack.com/product/index.jsp?productId=12651122

NTE176 (good) http://www.nteinc.com/specs/100to199/pdf/nte176.pdf

The NTE158 could be a good choice and Radio Shack can probably ship them to the UK. It's in a TO-1 package and Radio Shack also have a suitable heatsink clip: http://www.radioshack.com/product/index.jsp?productId=12673806
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Yes, right.

That's interesting about the ): getting included in the URL. Must be a bug in the forum software. It parses URLs automatically and usually gets them right.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
Heh would u look at that... I just noticed the latest version of the circuit simulator has Schmitt's.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
Turns out plccenter ship the NTE176 from the us and they have a minimum order of £35 any other ideas before I splash out?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Well, the obvious solution is to power the unit from four 1.5V cells instead of three.

Another solution would be to use a TI TPS72301 low-dropout negative regulator, which is available from Farnell UK http://uk.farnell.com/texas-instruments/tps72301dbvtg4/ic-ldo-regulator-0-2a-10v-sot23/dp/1207336 for £2.58, which has a dropout voltage of about 0.125V (typical) at 100 mA load. It's only available in a SOT-23-5 SMT package though. Let me know if you want to go that way and I'll show you how to connect it.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
Its not really the dropout or max power output I'm concerned about its the heat and efficiency.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
If you want to be able to provide up to 4.0V into those motors from a 4.5V power source, the minimum voltage drop you can achieve in the emitter follower is the important factor. That's why you suggested using a germanium transistor, and that's why I suggested a negative low-dropout regulator as an alternative since the minimum order value on the germanium transistors is prohibitive.

The amount of heat energy dissipated by the emitter follower (or whatever replaces it) is defined by the current through it, multiplied by the voltage across it. Neither of those is affected by its "efficiency". For a given motor speed, you have a certain amount of power that must be dissipated in that part of the circuit. That means that it will be converted to heat energy. You need to make sure the device won't get too hot. A heatsink may be needed.
 

CommanderLake

Oct 2, 2012
227
Joined
Oct 2, 2012
Messages
227
Hey is there any way to hook up an NPN along side the emitter follower so it enters saturation when the PNP is fully on so I can get a higher max output at which point efficiency is also increased?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Not easily, not that I know of. The closest option is a negative LDO regulator.
 
Top