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.

555 and BC327/BC337 complementary pair

Status
Not open for further replies.

boylesg

Advanced Member level 3
Joined
Jul 15, 2012
Messages
995
Helped
5
Reputation
10
Reaction score
5
Trophy points
1,298
Location
Epping, Victoria, Australia
Activity points
11,537
How do you get this sort of circuit to work?
555CompPair.jpg

I don't understand what I am doing wrong.

A complementary pair like this works perfectly well as part of a class AB amplifier but there seems to be no way to make it work from 555.
It appears that the 555 cannot switch the BCs on enough for them to generate or sink enough current from their emitters.

I have had this sort of problem before with a Jacob's Ladder kit, from Jaycar, that simply did not work once soldered. It also was driving a single BC327 which in turn drove a automotive power darlington.

- - - Updated - - -

It works just as well without the diodes, but I can only get it to output enough current to drive a LED directly, both to GND and from Vcc.

But what I don't understand is why it seems impossible to get enough current out of the totem pole to light up an automotive light bulb....even dimly.
 

BradtheRad

Super Moderator
Staff member
Joined
Apr 1, 2011
Messages
14,009
Helped
2,789
Reputation
5,576
Reaction score
2,696
Trophy points
1,393
Location
Minneapolis, Minnesota, USA
Activity points
104,496
A safety resistor is needed, otherwise there is nothing to stop severe spikes from going through the led's.

For the fun of it, I'm running a simulation.



If you put the NPN on top and PNP on the bottom, then no resistor or diode is needed to limit bias. Nevertheless it's a good idea to include the resistor, for safety.

Larger capacitors will lengthen the current spikes, as well as slow down the action so your eyes can follow the led flashes.
 

Orson Cart

Advanced Member level 3
Joined
Jun 18, 2009
Messages
800
Helped
347
Reputation
1,146
Reaction score
558
Trophy points
1,373
Location
New Zealand
Activity points
5,973
Yes the 680nF are too small, and you may not see the brief & small flash, hence 100uF suggested above, also your power source must be better than a flat 9V battery...
You need to have near 50% duty cycle on the o/p to keep the caps charged to half rail, else the mid point of the caps will shift...(luckily you have this - 50%)
 

boylesg

Advanced Member level 3
Joined
Jul 15, 2012
Messages
995
Helped
5
Reputation
10
Reaction score
5
Trophy points
1,298
Location
Epping, Victoria, Australia
Activity points
11,537
A safety resistor is needed, otherwise there is nothing to stop severe spikes from going through the led's.

For the fun of it, I'm running a simulation.



If you put the NPN on top and PNP on the bottom, then no resistor or diode is needed to limit bias. Nevertheless it's a good idea to include the resistor, for safety.

Larger capacitors will lengthen the current spikes, as well as slow down the action so your eyes can follow the led flashes.
I got it working with a totem pole comprised of 2SC4793 and 2SA1837 - the light bulb lights up dimly.

I tried 690nF and 470uF caps but it makes no difference to the max brightness of the globe.

I still don't understand why it is so hard to get any decent power out of it.

According to the data sheet a 100mA base current should give about 1A output from the transistors, and 100mA is well within the capability of a 555.

Even if I leave out the base resistor from the 555 output it makes no difference to the brightness of the globe.

So is there any way possible to get max brightness from the globe using this sort of arrangement?
 

Vbase

Full Member level 6
Joined
Apr 7, 2015
Messages
367
Helped
74
Reputation
148
Reaction score
72
Trophy points
28
Activity points
1,997
It wont work with capacitors.
Link D1, D2, R3 they aren't needed, connect 555 pin 3 directly to the 2 bases of the transistors.
Connect the 2 emitters directly to the LEDs, no series resistor.
Replace the caps of 680nF with 2 resistors of 1K.

If you want more light reduce the value of 1K resistors.
If you want to light up small bulbs connect one side of each bulb to the emitters and the other side of one bulb to ground, the other side of the other bulb goes to +12V. If you wish I can send you circuit.
It can be done without the transistors because the 555 gives enough current to drive LEDs or 1W bulbs.
 
Last edited:

Audioguru

Advanced Member level 5
Joined
Jan 19, 2008
Messages
9,218
Helped
2,129
Reputation
4,252
Reaction score
1,957
Trophy points
1,393
Location
Toronto area of Canada
Activity points
57,921
The 470 ohm resistor limits the current to 2V/20mA LEDs, if it is used with 12V/100mA incandescent light bulbs then of course the light bulbs will be VERY dim.
 

boylesg

Advanced Member level 3
Joined
Jul 15, 2012
Messages
995
Helped
5
Reputation
10
Reaction score
5
Trophy points
1,298
Location
Epping, Victoria, Australia
Activity points
11,537
It wont work with capacitors.
Link D1, D2, R3 they aren't needed, connect 555 pin 3 directly to the 2 bases of the transistors.
Connect the 2 emitters directly to the LEDs, no series resistor.
Replace the caps of 680nF with 2 resistors of 1K.

If you want more light reduce the value of 1K resistors.
If you want to light up small bulbs connect one side of each bulb to the emitters and the other side of one bulb to ground, the other side of the other bulb goes to +12V. If you wish I can send you circuit.
It can be done without the transistors because the 555 gives enough current to drive LEDs or 1W bulbs.

Actually the bulb is dim because the duty cycle is about 50% and pushing the duty cycle up quickly renders those caps useless, i.e. they wont work with DC or close to it.

I have been trying to measure the peak current through the light bulb with my multimeter - it has a setting where it holds a peak current measurement. But perhaps that function just doesn't work that well with a square wave? All I have managed to measure with it is 10mA but an actual current of 10mA is highly unlikely to result in the light bulb visibly lighting up....which was the case when I was measuring that 10mA.

The idea here is to create a viable FET gate driver by the way, until the TC4422s I ordered arrive though the mail.
 

Vbase

Full Member level 6
Joined
Apr 7, 2015
Messages
367
Helped
74
Reputation
148
Reaction score
72
Trophy points
28
Activity points
1,997
Are you saying that you wanted to keep us busy until your bits arrive?! very thoughtful.
 

boylesg

Advanced Member level 3
Joined
Jul 15, 2012
Messages
995
Helped
5
Reputation
10
Reaction score
5
Trophy points
1,298
Location
Epping, Victoria, Australia
Activity points
11,537
Are you saying that you wanted to keep us busy until your bits arrive?! very thoughtful.

Well no, I was genuinely trying to figure out why the current appeared to be so damn low because, if it was, then it wont switch FETs on properly.

I am still unclear why the max current function on my multimeter does not work particularly well in this context but it seems to work perfectly well measuring the peak current flowing through a flyback primary.

I have not had a multimeter with this sort of function on it so I have no idea what the limitations of it are.

Are you able to enlighten me here?

I have been told that the frequency measurement functions are not always accurate on multimeters, although with this multimeter, it seems pretty good. It also has duty cycle and it also seems fairly accurate.
 

Orson Cart

Advanced Member level 3
Joined
Jun 18, 2009
Messages
800
Helped
347
Reputation
1,146
Reaction score
558
Trophy points
1,373
Location
New Zealand
Activity points
5,973
If you make the base resistor zero ohms (no diodes either) and place your lamps across the transistors you will get the bright lights you desire...
 

Vbase

Full Member level 6
Joined
Apr 7, 2015
Messages
367
Helped
74
Reputation
148
Reaction score
72
Trophy points
28
Activity points
1,997
If you make the base resistor zero ohms (no diodes either) and place your lamps across the transistors you will get the bright lights you desire...

This is what I've already advised boylesg. Please read his embarrassing response that followed, post #7.
 

boylesg

Advanced Member level 3
Joined
Jul 15, 2012
Messages
995
Helped
5
Reputation
10
Reaction score
5
Trophy points
1,298
Location
Epping, Victoria, Australia
Activity points
11,537
This is what I've already advised boylesg. Please read his embarrassing response that followed, post #7.

Done all that except that putting the globe directly across the transistors does not work. I suppose because when the NPN is turned on the PNP is turned off so the current has no where to flow to.

What does work is to connect the other end of the lamp to either Vcc or GND. Then I get full brightness and good control of the brightness through the duty cycle on the 555.

I should try my multimeter in max current mode and see if it can measure the peak current properly in this situation.

I ended up getting 2 nice darlington transistors so that the 555 is not taken close to its limits.

One thing that stumps me is how they manage to get very close to Vcc output from those gate drive ICs. With the darlingtons I get Vcc minus around 2V on the output.

By the way I discovered that one the transistors I started out with was fried and accounted for why I could not get this to work at all to begin with - doh!
 

Audioguru

Advanced Member level 5
Joined
Jan 19, 2008
Messages
9,218
Helped
2,129
Reputation
4,252
Reaction score
1,957
Trophy points
1,393
Location
Toronto area of Canada
Activity points
57,921
One thing that stumps me is how they manage to get very close to Vcc output from those gate drive ICs. With the darlingtons I get Vcc minus around 2V on the output.
The datasheet for the 555 shows that its output has an old fashioned TTL darlington transistor emitter-follower to pull up its output to about 1.0V to 2.5V less than its supply voltage. The datasheet for the modern TC4422 Mosfet driver IC shows that its output has Mosfets that switch very close to the supply and to ground.
 

SunnySkyguy

Advanced Member level 5
Joined
Sep 26, 2007
Messages
6,743
Helped
1,675
Reputation
3,348
Reaction score
1,644
Trophy points
1,413
Location
Richmond Hill, ON, Canada
Activity points
50,732
How do you get this sort of circuit to work?
View attachment 116746

I don't understand what I am doing wrong.

A complementary pair like this works perfectly well as part of a class AB amplifier but there seems to be no way to make it work from 555.
It appears that the 555 cannot switch the BCs on enough for them to generate or sink enough current from their emitters.

I have had this sort of problem before with a Jacob's Ladder kit, from Jaycar, that simply did not work once soldered. It also was driving a single BC327 which in turn drove a automotive power darlington.

- - - Updated - - -


It works just as well without the diodes, but I can only get it to output enough current to drive a LED directly, both to GND and from Vcc.

But what I don't understand is why it seems impossible to get enough current out of the totem pole to light up an automotive light bulb....even dimly.

Many flaws with this design.
Let me clue you into reasons.
  1. D1,D2 are reversed for linear Class A,B cct and need pullup/down to bias Vbe drop, instead you are using them as logic steering diodes for bipolar LED switch.
  2. All LEDs ( with single junction) are rated for -5V ABSOLUTE Max. You are applying up to -10Vpk on start up and which may cause catastrophic failure. ( dim or dead) Never do this.
  3. You are AC coupling low ESR LEDs and all diodes and semiconductors are low when saturated thus short RC=T duration.
  4. In fact you can estimate it easily using
    mimetex.cgi
    1Ω per Watt rating so W*ESR
    mimetex.cgi
    ~1 at 25'C only, ESR drops with rising temp. (Stewart's theorem)
    e.g. 5mm LEDs are around 15Ω * 65mW ( ~=1) 3W LEDs ~=1/3Ω
  5. If you want to AC pulse LEDs, use that for RC ~Time Constant.
  6. But you can compute it from datasheet , from Vol/Iol and (V+ - Voh)/Ioh
  7. Running 1 LED off 12V is very inefficient. If Red/Yel assume 2.2V if Blue/White assume 3.2V nom. then add more in a string and use Ohm's law for difference. current limiter.
  8. Learn to interpret diodes when saturated as Vth+ If*ESR. For LEDs Vf always increases with decreasing wavelength due to chemistry and energy of photon at threshold voltage Vth and increases with If to yield rated Vf @ If.

Tony Stewart

Also when running a string of LEDs that are identical, you can put assume Vr is plit evenly. But if not matched, then you need back-back diodes like you have done. If you plan on a regulated voltage or good battery, choose the Rdrop for current limit based on Vdrop <= 1 Vf for efficiency. THen include ESR of driver + all diode drops in Rdrop for current limit.

Then you can use different colors like BLue/Red and change duty cycle to change ratio of colours used using Pot for PWM.
Any Astable can do this even any Schmitt trigger gate.

Now show us your redesign.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top