Good
Ways of Blowing up NPN Transistorized Controllers Updated December 2013 |
||||||||||||||||||||
I guess most of the readers will have one of the 3 most common types of controller circuit (PNP, NPN, and switching) and they'll be mainly interested interested in the type they own. To aid clarity I've written these tips on 3 separate pages - one for each of the 3 most common types of controller circuit . This is the one for the "NPN" circuit commonly used in imported controllers, much of the content is the 2 other pages is similar but is specific to the PNP circuits, and switching controller circuits | ||||||||||||||||||||
1) Shorting Out The Tapes Another how it works question,,,, what happens if there's a short circuit across the motor (e.g. the chassis shorts out the tapes)? Answer - The transistor tries to
provide more current to compensate. A short circuit could well force 100
amps through the transistor, or provide 300 + watts of
heating. OK, maybe 99 times out of a hundred it won't be that bad - but
you don't want a blown transistor on the hundredth time! What does this mean? Answer - If you get a good solid
short, the transistor your transistor might fail - you might get away with instantaneous shorts,
but these could cause some damage to the transistor which will cause odd
characteristics and perhaps failure next time the transistor gets a little warmer
than normal. How do you get round this problem? Answer - Put some sellotape or varnish on the bottom of you cars so they don't short out the tapes. Always unplug your controller (or at the very least don't press it) when somebody is putting the tapes down - if they are using something solid and metal it will produce a very good short circuit. If you are repairing the tapes please take a care to warn the driver! (Quite apart from any controller damage, it'll stop them driving their cars into you!) |
||||||||||||||||||||
2) Other Good Ways of Producing Short Circuits Something which has always surprised me is that lots of transistors are bolted direct to the heatsink so the whole of the metalwork is connected to the collector terminal - the tape + on the track. This is all very well as long as it doesn't come into contact with (for example) the negative tape , the brake terminal on a controller plugged into the next lane or the fixing screw of a controller socket*. This sort of thing causes the transistor to fail (often to dead short which will put your car straight into the wall). This is not a problem that occurs very often, but it could be prevented with a transistor mounting/insulating kit at a cost of a few pennies. There is a small disadvantage to insulating the transistor. Electrical insulation will also increase the thermal resistance between the transistor and heat sink. This means that the transistor will run slightly hotter in order to transfer the same amount of heat. If you use heat sink compound and bolt down the transistor properly (do the screws up tight) the temperature difference should be small. Higher temperature = less reliability, if you think this small temperature rise due to insulation will make any difference I would recommend you fit a slightly larger heat sink and insulate the transistor - or better still a much larger heat sink and insulate the transistor. * Modern controller sockets have the fixing screws connected to the E terminal (the negative power supply on a slot track). |
||||||||||||||||||||
3) Some Like It Hot!! (but your transistor doesn't) How hot does it get - well this circuit produce just as much heat as a variable resistance controller (a diode controller also produces the same amount of heat). Transistors are easy to mount on heatsinks to get rid of heat ,,, and a good thing too because a controller resistor can run at much higher temperature without failure - also a controller resistor often starts to smell before it fails (so there is time to change before it blows or stinks out the clubroom) but a transistor will just fail. To avoid transistor failure you must stop it getting too hot - the only way to do that is to mount it on a big enough heat sink. As a rule of thumb (or finger?), if after a couple of minutes running the transistor is gently warm to the touch its OK - if it's hot to the touch (50 degrees or above) then a larger heat sink would help reliability and if its too hot to keep your finger on, (70 degrees C or above) then unplug it quick! - it either has a fault or it needs to get rid of more heat - try a larger heat sink or a fan. Transistor manufacturers typically recommend about 20% reduction in current if the case is running at 50 C and around 40% reduction at 75C . If you do need a bigger heat sink, probably the easiest way of getting something effective is to buy a larger heat sink from one of the electronics suppliers. These days a heat sinks intended for computer processors are widely available, and are very suitable. However heat sinks don't have to be a length of finned aluminum - anything that helps get rid of heat will do - the metal box bolted to the finned aluminum will increase it's effectiveness. Also some heat sink compound will help a little. Why can't I tell you what size heat sink you need? Well, the amount of heat a controller produces depends on many things, a big enough heat sink for all possible circumstances would be far too large for most people. Things which make controllers run hotter are;- Inexperienced drivers, |
||||||||||||||||||||
4) The Dirt Bits! Probably the most frequent cause of transistor failure is tracking between the brake terminal and the top of the resistor contacts. This is caused by a conductive deposit on the contact board where the wiper runs - usually remains of the contact brush etc. Initially this causes the transistor to get warmer than normal (because it is partly switched on when the controller is on Brake), if not corrected this causes the transistor to get hotter until it fails. Many people have noticed the transistor getting warm and solved the problem by cleaning the tracking off the board before it causes a serious problem. The areas marked in red on the diagram are the important areas to clean.
You can
avoid this problem by cleaning the board every couple of months as a matter
of routine maintenance. Some controllers are very much more prone to this problem
than others. The deposit builds up much more quickly when copper/carbon brush
material is used as the wiper and it also depends on the gap between the brake
band and the top of the resistor. Incidentally, if you are wondering why this
was never a problem on variable resistance controllers (it shouldn't be a problem
on Diode controllers either), the answer is that the tracking probably does
occur, but the tiny currents that pass through the tracking are negligible to a
slot car motor - but they are multiplied to by the gain transistor. This tracking is less of a
problem with this circuit than the "normal British Darlington" circuit because
the single transistor only multiplies the tiny current by about 100 where as the
Darligton pair used in a "normal British" multiplies the tiny current
by 1000+. |
||||||||||||||||||||
5) A Bridge Too Far Another problems that can cause transistor failures - not to mention damage to variable resistance controllers - is the wiper bridging the gap between the brake and the resistor (even momentarily). Make sure the wiper cannot touch the resistor contact and the brake at the sale time - it is much safer if there is some "Dead" wiper travel (that is after the wiper leaves the brake there should be some travel when it's not connected to anything before it makes contact with the resistor). This circuit has two electrically isolated wipers / contacts connected to the trigger. Make sure they stay electrically isolated from each other. Where controllers have separate brake contacts, power contacts, or full power relays make sure they work in the correct sequence. Generally this is how the should work (starting with the trigger / plunger in the fully released position)
The controller must follow the sequence in reverse when releasing the trigger / plunger from the full power position Note microswitches and relays often don't release at the same point they make contact, so you need to check the sequence in both directions.
|
||||||||||||||||||||
6) Track Wired Incorrectly Transistors only pass current in one direction - if the track is wired to BSCRA standards (with the controller between the + side of the power supply and the track) the normal circuit works fine. If the track is wired incorrectly (with the controller between the negative side of the power supply and the track) your transistor will blow up. Incidentally, this has got nothing to do with wiring the car up to go forwards or backwards. Correct Track Wiring (updated to 2003 BSCRA Standards) |
||||||||||||||||||||
The
preferred solution
to this is to get the track wired up properly. If it isn't an option, a variable
resistance controller will work quite happily which ever
way the
current passes. However, the rest of the country's
transistorized controller owners will still have their controllers wired for
normal track wiring, so unless you are really dedicated to discouraging
new members who have moved from other clubs and visitors, it really is a
good idea to wire the track up the standard way round.
If you want a transistorized controller with the "NPN" circuit to work the other way round, a MJ14003 PNP transistors will work. As I've mentioned earlier, I haven't managed to find out what NPN transistors these manufacturers use, so I cannot comment on how similar this PNP equivalent is to the original transistor . |
||||||||||||||||||||
7
Diodes
Some controllers (e.g. Pro 2 and Pro 3) have a diode connected to the transistor. The diode is reverse biased in normal operation (or to put it another way, no current flows through the diode when the controller is correctly connected.). The diode will provide some protection if the controller is connected up incorrectly, or it is plugged into an incorrectly wired track. In these incorrect conditions, the diodes can pass high currents, and these diodes have been known to fail. When they fail the controller is likely to give more or less full power all the time (even without pulling the trigger), much the same symptoms as some sorts of transistor failure. If the diode has failed, simply disconnecting one end of the diode will solve the problem but it will remove the protection against incorrect controller connection etc. The replacement diode should be a schottky rectifier diode, I'll not recommend a particular type, most electronics suppliers have dozens of types rated at more than 20 amps and more than 40 volts. I haven't proved this, but it might improve reliability to go rather higher than 20 amp rating; there is little cost penalty in going to at least 40 amps. (If you are interested in an explanation of that, it is difficult to predict what peak current the diode will see for a short time in fault conditions, although 20 amp rectifier diodes are usually designed to take at least 100 amps peak current.) |
||||||||||||||||||||
If you do ever blow up a transistor on one of these controllers, I suggest you buy the replacement from the supplier. I could suggest alternative transistors that look good on paper (ones with high current and heat dissipation), but I won't recommend anything till somebody has tried it. (Another couple of catches are that these transistors are only from professional electronics stockist (retail stockist don't seem to keep this sort of thing any more), and aren't any cheaper than the replacement from the controller supplier.) |
Related pages -
Copyright © 2009, 2013 C.Frost All rights reserved
No liability is accepted for the information on this site or any use to which it may be put.