Discussion in 'Electronic Projects Design/Ideas/Reviews' started by fezder, May 14, 2012.

1. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
hmm, well schematic would help i experienced a littlebit on my own, luckily didnt burn anything, but i managed to adjust current somewhat, by using fet and potentiometer as voltage adjustment to fets gate. but i had to add limiting resistor, and fet gets really hot, and i can only lower the current, not rise it

2. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hi,

You really do need a heatsink on the transistor whatever kind you use. It's being operated in the linear mode so it's going to get hot.
I'll draw up a schematic either later or tomorrow sometime. Not too much too it really.

Check out post #80 i attached it to that post by mistake.

Last edited: Jun 20, 2012
3. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
ah, thanks, that should help just to make sure that i understand how circuit works: op amps output voltage and current vary from reference voltage given to positive input of op amp, which therefore controls current flow throught transistor, and resistor at transistors emitter is current limiter? could you correct me if im wrong, its always better to understand how something works, not only know that it works?`

Last edited: Jun 21, 2012

Joined:
Jan 12, 1997
Messages:
-
Likes:
0

5. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ

Hi again,

The resistor at the transistor emitter senses the current level. It converts the current into a voltage. That voltage then appears at the op amp inverting terminal. The op amp puts out a voltage that drives the transistor such that when the voltage at the inverting terminal equals the voltage at the non inverting terminal the op amp stops changing its output voltage. So the op amp tries to drive the transistor in a manner that will make the inverting terminal voltage equal to the non inverting terminal voltage.
When it accomplishes this, the current through the emitter resistor is approximately equal to the current through the collector and so that sets the collector current which is the load current.
The emitter resistor is typically small like 0.1 ohms but it could be higher like 1 ohm. You could also include a collector resistor to take up some of the power because the transistor ends up dissipating all the power. And of course the transistor needs a heat sink.

Also, since the output current level may be high you might need an additional driver transistor to drive the bigger transistor. That driver transistor can be smaller than the larger one. The connection would be base to op amp output, collector to big transistor collector, and emitter to big transistor base. When the smaller transistor conducts it pumps current into the base of the big transistor and that provides for more drive. The op amp output is limited to maybe 20ma max, and depending on the gain of the output transistor that may not be enough to drive the load so the smaller transistor helps provide more current into the base of the larger transistor.

6. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
7. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hi,

Yes you can do it that way.

8. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
ok, just one more stupid question....what should be used as thermal conductor, and for electric insulation that heatsink doesnt come live from collectors voltage? i dont have to-3 case install kits

9. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hello,

Well if you use some kind of insulating stand offs for the heat sink you dont have to isolate the transistor case from the heat sink.
If you do want to isolate, you have to find si pads or use mica and thermal grease. Even without mica use thermal grease.

What i do for my test sets is i use a small piece of wood and mount the heat sink on that with bolts and nuts. I then mount the
circuit board on a different part of the wood using standoffs and bolts. A few small feet on the bottom and it works nice. Just
have to be careful about the bolt heads under the wood, so sometimes i counter sink those so they cant touch anything metal if
the wood is put on top of something metal. A small pine wood board works pretty well. 3/4 inch thick wood works best, but i've used as thin as a 1/4 inch piece for a microcontroller board with no problems. This is for the lower voltage stuff as the higher voltage (400 volts or like that) might conduct too much through the wood.

Last edited: Jun 21, 2012
10. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
hmm, i dont know why but it doesnt work, i just hook load points of schematic to resitors place in circuit? or is it because i used potentiometer as voltage reference at positive input of op amp

11. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hi,

What doesnt work? What circuit are you using and what do you have it connected to?

12. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
oh sorry, i mean that previous schematic which you posted, load tester at post #80. i have same connection as in schematic, except resistor before potentiometer which is connected to load, and i have total total circuit connected to coils load, as in place of resitor. see link, i hope it makes sense, paint you see :S
http://i45.tinypic.com/zxmlud.gif

13. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hi,

That drawing looks ok. There's a catch though using only one transistor. The op amp can only ouput maybe 20ma, and if the gain of the transistor is 50 that means the max collector current can only be 1 amp (20ma times 50). That means that if you try to get more than 1 amp it may not work so you'd need another transistor to drive the bigger one.

You can also check for oscillation just to make sure there is none.

14. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
well i drive tip122 darlington with op amp, and tip122 drives BD182, pretty similar to 2n3055? how do i check oscillation? the op amp im using is LM358 see link, i took a photo from breadborad assembly
that capacitor is c1 at drawing, had to change it because i changed power supply to 19v and therefore 16v cap wasnt sufficient, thats 25v
http://i49.tinypic.com/18y6hu.jpg

15. ### 4pyrosWell-Known MemberMost Helpful Member

Joined:
Oct 22, 2010
Messages:
4,368
Likes:
282
Location:
Pennsylvania, USA
You should not put high power through the breadboard.

16. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
hmm, yes you are right, i just started making this to breadboard, that it would be easy to change values :S

17. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hi,

Oh you are still using that breadboard? That's only good for about up to maybe 1 amp. This should be soldered. Also, you need to watch the wire sizes depending on current level.

If you use three transistors you might get too much voltage drop from collector to emitter of the final transistor. You have to watch for that too. If the normal sat comes out to 3 volts then you will never be able to pull the load lower than 3 volts which might mean you can not get to a high enough current. Perhaps you can check it with 5 volts on the output and see what current you can get.
One way or another it should be able to work.

You normally check for oscillations with a scope. You look to see if there is any wild variation at the collector of the final transistor. There might be some variation because of the buck circuit, but it should not be too great.

18. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
hmm, well i solder them, i newer thought about breadboard limits. i checked oscillations and couldnt find anything big, only wery small fuzz. i hesitated too much and didnt take photo from scopes wief :S. i managed to change current with fet, which supplies small amount of current to base of transtor, should i stick with this as now i can change current, or should i stick to your idea, your idea is probadly better as your experience is far better than mine?

19. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Can you post the circuit?

20. ### fezderWell-Known Member

Joined:
Dec 11, 2011
Messages:
1,665
Likes:
100
Location:
Mikkeli, Finland
yeah, sure, heres the schematic, i used 2n3055 as example, maybe i should drive it with smaller transitor thought, as it would stress fet less?
http://i46.tinypic.com/1zml98h.png

21. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
Hi again,

Well you really have to show the values of the components. A MOSFET might be harder to control in this config, but it might work to some degree.