Darlington transistors, more bang for the buck.

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
perfectbite
Posts: 255
Joined: Tue Jan 07, 2003 1:01 am
Location: California
Contact:

Darlington transistors, more bang for the buck.

Post by perfectbite » Thu Jun 24, 2004 11:24 am

In a reply to honkkongphooey Stephen wrote:<p>"Four darlington transistors will get the current level up high enough to drive the stepping motor coils."<p>So Darlington transistors can do more than just be the equivalent of electro/mechanical latching relays?<p>
In thinking of it I'm starting to be able to answer my own bloody questions here.

User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: Darlington transistors, more bang for the buck.

Post by Chris Smith » Thu Jun 24, 2004 2:38 pm

Darlingtons are merely large levers. <p>Think of a one foot crow bar, VS a three foot long crow bar. <p>Same work is done, but less force is applied, for a longer stroke, to get the same nail pulled. <p>A small signal in or a tiny signal in, is sensitive enough to cause the cascade effect inside the transistor to deliver a larger signal current out. <p>But time is sacrificed for this advantageous action. Saturation of the transistor takes longer to turn on and off.

User avatar
haklesup
Posts: 3006
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: Darlington transistors, more bang for the buck.

Post by haklesup » Thu Jun 24, 2004 6:15 pm

In a Darlington pair transistor, the first transistor is used to drive the base of the second. The result is greater gain than a typical bipolar transistor. <p>If each transistor has a gain of 100, a darlington pair of them yeilds a gain of 1000. You can build it out of seperate transistors but having them in the same package has advantages of matching the electrical and thermal characteristics.

dyarker
Posts: 1744
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Re: Darlington transistors, more bang for the buck.

Post by dyarker » Thu Jun 24, 2004 6:31 pm

A Darlington connection is two transistors. Often a power transistor (A) with a small hfe (gain) connected to a smaller transistor (B) that has a higher hfe.<p>The connection is collecters A and B connected together, emitter B to base A, base B for input, and emitter A.<p>The gain of the connected transistors is hfeA times hfeB. But the forward base-emitter drop is around 1.4V for silicon because it is two base-emitter junctions in series.<p>In a Darlington transistor the manufacturer puts the transistors in a single package with three leads. Collecter AB, emitter A, base B.<p>They were useful for motor controllers and heavy relay drivers, but power FETs do that job better now.<p>Does that help? (and agree with Chris' and haklesup's comments)<p>[ June 24, 2004: Message edited by: Dale Y ]</p>
Dale Y

Vincent
Posts: 45
Joined: Wed May 12, 2004 1:01 am
Contact:

Re: Darlington transistors, more bang for the buck.

Post by Vincent » Thu Jun 24, 2004 9:10 pm

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by haklesup:
In a Darlington pair transistor, the first transistor is used to drive the base of the second. The result is greater gain than a typical bipolar transistor. <p>If each transistor has a gain of 100, a darlington pair of them yeilds a gain of 1000. You can build it out of seperate transistors but having them in the same package has advantages of matching the electrical and thermal characteristics.<hr></blockquote><p>I thought that you multiplied the gains of the two to come up with the overall gain of the darlington?

rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: Darlington transistors, more bang for the buck.

Post by rshayes » Thu Jun 24, 2004 9:30 pm

Power FETs can be used to drive stepping motors, but they are one of several choices. The most common power FETs require around 10 volts of gate drive. This means an additional level shifting stage when operating from 5 volt logic. The darlington requires about 1.4 volts. Even TTL, which operates with a high level of around 3 volts, can drive this, possibly with the aid of a pull up resistor. In many cases, the logic can supply the milliamp or so of current without extra buffering.<p>Darlingtons are slow, with switching times in the 10 to 100 microsecond range. Power FETs apppear to be fast, with switching times in the tens of nanoseconds.These measurements are often made with a signal brought in with a terminated 50 ohm cable. The effective impedance driving the gate is 25 ohms. Developing 10 volts across that terminating resistor requires 200 milliamps. Most practical drive circuits are not this robust.<p>The gate-drain capacitance of a power FET is usually in the 1000 to 10,000 picofarad range. This will slow down most logic circuits, so a buffer stage of some kind will practically be required. A power FET driven by a CD4000 series logic gate will probably be as slow as a darlington transistor stage.<p>Cost may be lower for the darlington transistor by a factor of two or three, depending on the actual parts being compared.<p>Actually, the cheapest approach might be a discrete small signal transistor operating as an emitter follower from a 5 volt supply driving the base of a standard power transistor. Small signal transistors such as the 2N2222 cost about a nickel and can provide about 50 milliamp of drive in this situation.<p>It all depends on what you really want and are willing to pay for.

perfectbite
Posts: 255
Joined: Tue Jan 07, 2003 1:01 am
Location: California
Contact:

Re: Darlington transistors, more bang for the buck.

Post by perfectbite » Thu Jun 24, 2004 10:51 pm

Thank you. I have two dependent questions.<p>Stephen's original statement was;<p>"Four darlington transistors will get the current level up high enough to drive the stepping motor coils."<p>".........will get the current level up........"<p>I realize that water flow is a very poor analogy but my first question is if I have a constant source of water in a sluice/trough that pours out of one end in a steady stream and have a quick action sluice gate (a switching transistor) at its outlet. When the gate is closed the water will build up in the trough and, when the gate suddenly opens the built up water will surge out and then fall back to its steady stream. Now going back to a circuit. If I open and close the gate quickly enough (via the Darlington or something like it) so that only the surges get by have I supplied 'more' Amps to my driven device downstream? Is that what he meant when he wrote - "will get the current level up"?<p>After the discussion on unity and over unity devices it became abundantly clear to me that a viable over unity machine could produce a completely useless few ergs so you can't just generate extra Amps out of thin air. You either get the current level up by providing more Amps to your circuit or by manipulating the Amps you already have. <p>So my second question is - If the sluice gate analogy holds true what SIMPLE property of a circuit is being utilized in 'getting the current level up'? That is if the sluice gate analogy holds water. (get it?) <p>I assume that Induction proper, belongs to loads and could, as it were, suck the Amps out of the switch but the Amp surge would have to be there to be sucked out.<p>I'm also assuming that Capacitance belongs with Voltage more than with Amperage. <p>
BTW I have (donated labour) fixed the local Habitat for Humanity chapter's power tools for the last 8 or 10 years and, because they are having their 'Summer of '04 Build-a-thon' in a few weeks, have been asked to check their power tools and extension cords before the big event. I can have five assistants who know a little or six assistants who know nothing. I chose to have the ones who know nothing. Some of their extension cords are so large that they cart them around in a wheelbarrow. Does any one have a simple trick or know of a gizmo for testing cold 110VAC extension cords. Hot I'll have to be there, cold, the assistants can feel they're really doing something while I do something else. There's a few small welding jobs too.<p>The foundations, roofs, plumbing and electrical are done by contractors but imagine having to deal with 150+ ready and willing volunteers singly and in groups who know very little or NOTHING about building houses or even about being on a building site turning up at 8 a.m. ready to work with their hammers and tool belts. My friend, who's the superintendent, has to have the materials and tasks ready for them ALL to do a full week end's work otherwise they'll feel they haven't accomplished anything. What a headache that job could be! He's good at it, he's been doing it long enough.<p>[ June 25, 2004: Message edited by: perfectbite ]</p>

bridgen
Posts: 196
Joined: Sun Aug 25, 2002 1:01 am
Location: Camberley, United Kingdom
Contact:

Re: Darlington transistors, more bang for the buck.

Post by bridgen » Fri Jun 25, 2004 6:10 am

There is one characteristic of the Darlington pair which hasn't been mentioned so far in this thread. <p>That is its higher Vce(sat) <p>A single transistor passing a hefty current can be expected to have a Vce(sat) of a few hundred millivolts to over a volt. <p>In a Darlington the figure will always be around 0.7V higher than in a single transistor because it is the sum of Vbe of the power transistor plus the Vce(sat) of its driver.<p>[ June 25, 2004: Message edited by: David Bridgen ]</p>

User avatar
haklesup
Posts: 3006
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: Darlington transistors, more bang for the buck.

Post by haklesup » Fri Jun 25, 2004 12:02 pm

The Sluce gate analogy only holds water if you consider the volume in the trough that the water builds up in, is Capacitance. <p>The capacity of your pipe to overfill (above average depth while flowing) is what gives you the surge current. The voltage pulse comes when the trough overfills and the resulting rise in (head)pressure causes an increase in current flow. The time it takes to recover to normal depends on how fast (switching time) and how far you open the gate (resisance)<p>In an electric curcuit, the electrons are generally confined more tightly to the wire than water in an open trough, a closed pipe at pressure makes a better comparison. <p>The plumbing analogy works in a variety of situations, you have to be careful how you interpret it though. Dynamics (Mechanics) also makes a good model, especially if you consider the math is often identical.<p>Capacitance is related to voltage and current equally because V and I are always related by the characteristics of the load (be it C, L R or a mix). When analyzing circuits you can consider either the V or the I signal component, whichever is convenient to understand and analyse at the time but both are always intimately linked.<p>10,000 yes, a typo, me bad :eek:

rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: Darlington transistors, more bang for the buck.

Post by rshayes » Fri Jun 25, 2004 1:47 pm

Bipolar transistors and power FETs both control the flow of current through an external circuit. Both devices have saturation regions, where the collector or drain voltage is so low that the device looses control of the current and it is determined by the external circuit.<p>When a bipolar transistor is unsaturated, the current flowing from the collector to the emitter depends on the current flowing from the base to the emittter. The ratio between these currents is called the current gain of the transistor (sometimes called beta). The current gain depends on the transistor construction and use. High voltage power transistors (above about 300 V)usually have betas in the 5 to 10 range. Lower voltage power transistors (40 to 80 V) usually have betas of at least 20. Small signal transistors usually have betas above 60 and in some cases as high as 300. Current gain is not a tightly specified characteristic. Usually, there is about a 3 to 1 range or so for individual transistors of the same type. The critical characteristic is the minimum gain that a transistor type is guarenteed to have.<p>The base voltage must be about .6 to .7 volts for base current to flow. Without base current, there is no collector current and the transistor is considered to be "cut off".<p>When the collector voltage falls to a low value, the collector current is no longer determined by the current gain. The current is limited to a value less than beta times the base current by the external citcuit.<p>In a power FET, the current from the drain to the source is controlled by the voltage between the gate and source. This ratio is called the transconductance and is the drain current divided by the gate voltage. It can vary widely, depending on the individual device and operating point.<p>Power FETs also saturate at low drain voltages. Again, the current is then controlled by the external circuit and is less than the tranconductance times the gate voltage.<p>Power FETs have a threshold voltage. No current flows until the gate voltage exceeds a certain value. This is usually around 3 volts, but can be higher or lower.<p>It is often easier to work backwards when designing a circuit. For example, if the stepping motor winding requires .5 amp, then the base current required to produce this current is .5 amp divided by beta. Beta is probably between 20 and 60, so the base current will be at least 8 milliamps and possibly as much as 25 milliamps.<p>This is more current than most logic gates can provide, so a second stage is necessary. In a darlington connection, this additional stage is an emitter follower with the emitter of the first transistor connected to the base of the second transistor. The emitter current of the first transistor provides the base current for the second transistor. The emitter current is the sum of the collector and base current, and is beta plus one times the base current of the first transistor. If the beta of the first transistor is at least 40, then its base current will be less than 25 milliamps divided by 40, or about 625 microamps. This is a level than can be provided by a logic circuit.<p>The voltage needed is the base-emitter voltage of the second transistor (about .7 V) plus the base-emitter voltage of the first transistor (another .7 V) for a total of 1.4 volts. This can be driven by a TTL gate, which has a high logic level of about 3 volts. A series resistor to limit the base current willl be needed. This will have a maximum value of 1.6 volts (3 V minus 1.4 V) divided by 625 microamps, or 2560 ohms. Use 2200 ohms. <p>One quirk of TTL is that it supplies much less current for positive going signals than for negative going signals. The base current for the darlington can be supplied by a pull up resistor of 2700 ohms between the logic gate output and the positive 5 volt supply. The logic gate no longer has to supply any current to turn on the darlington, but it must divert about 1.8 milliamps to ground to turn the darlington off. Most TTL gates are easily capable of this.

perfectbite
Posts: 255
Joined: Tue Jan 07, 2003 1:01 am
Location: California
Contact:

Re: Darlington transistors, more bang for the buck.

Post by perfectbite » Sat Jun 26, 2004 8:49 am

Thank you for sharing your knowledge. I apologize for my poorly worded questions. There's my cogitative meal for the next little while. I am incredibly dense (thick) but with your very valuable input I have come to appreciate that from the basic Ohm's Law the various permutations, aspects and ways of managing those basic E/I/R quantities are worthy of study in and of themselves. Even to the point that some significant aspects of those functional relationships have been quantified using their discoverer's names.<p>(I sound like the preface to an Intro to EE 101)<p>Thank you all again for your kindness and patience.

toejam
Posts: 253
Joined: Wed Dec 05, 2001 1:01 am
Location: n.c.
Contact:

Re: Darlington transistors, more bang for the buck.

Post by toejam » Mon Jul 12, 2004 6:21 pm

there are a lot of logic level mosfets out there ckeck out parts houses. I think they are easier to design with than darlingtons or any other type of transistor.The only thing you have to remember is to turn them off as well as on.

Post Reply

Who is online

Users browsing this forum: No registered users and 12 guests