CandOfan

In another thread the concern was raised about relatively inefficient motors drawing current sufficient to heat up and possibly cooking a decoder.

If this is a potential problem, what can one do—other than remotoring—to conduct the heat away from the decoder? In a PC the usual answer is a bigger heat sink. Is it reasonable to, say, thermally attach the decoder or the most important chip in the decoder to a piece of metal? Or even, in the case of a brass model, potentially to the model itself?

Whilst I'm quite aware of the general principle, I have honestly never heard of decoders thermally destroyed unless their current limits were exceeded. If this is a common thing, why don't decoders have heat sinks? Or at least heat sink options? I find it curious that even the O/G scale decoders, which handle 4+ amps, do not have heat sinks.

Modeling the C&O in Virginia in 1943, 1927 and 1918

Reply 0
Prof_Klyzlr

Heat Dissipation

Dear C&O,

When it comes to electronics, heat dissipation is always "a thing"...

...however, specifcally in regards decoders, the devil is in the details...

In no particular order of preference:

- When it comes to calc-ing the "power handling" of decoders, there is occasionally some "mis-understanding" about how the values are reached. Sometimes the stated power rating of a decoder includes both the Motor circuit and the cumulative loads of the (how many?) function outputs, sometimes it's calc'd only on the "Motor Output". Keep eyes open, and make sure to Read the Specs...

- Older, and certainly smaller decoders tended to run with components which had little/no Electrical headroom. The original TSU-750 "micro" sound decoder comes to mind, was commonly noted as benefiting from being mounted against the cast-metal frame, or on a brass-sheet "heat spreader"/"heatsink"...

http://members.optusnet.com.au/mainnorth/750Temp.htm

https://tonystrains.com/news/heat-sinking-the-soundtraxx-micro-tsunamis/

https://tonystrains.com/news/heat-sinking-the-soundtraxx-micro-tsunamis-february-12-2010/

https://www.trainboard.com/highball/index.php?threads/heat-reflection-for-dcc-sound-decoders.74121/

NOTE! this is NOT the same as smearing "arctic silver" directly on the surface of the decoder uPC or H-bridge transistors and bonding to a lump of metal. Rather, keeping the factory-stock heatshrink intact and just mounting the decoder heatsink surface flush to the frame (held in place with Kapton tape) is enough.

- Flipside, with a little bit of planning, it's actually possible to mount a decoder, particularly in HO diesel models, such that the decoder "floats" above the rear truck, with a signficant ammount of free-airspace around it in which to "vent to atmosphere"... Mounting in a steam tender would similarly allow "airspace" mounting.

If I may quote from the current ST TSU2 Installation manual
https://soundtraxx.com/content/Reference/Manuals/Tsunami2/Installation-Guide.pdf(page 8)

Quote:

When  planning  a diesel  installation,  we recommend  providing  additional  airflow  if  possible. When possible avoid  mounting  the  decoder  above the  motor;  the  motor  will  raise the  decoder’s temperature during  operation. 

- The good news is that over time and component/decoder developement,
higher-power-capable components have become physically smaller, 
the decoder manuf's have developed more-efficient Motor-drive circuits which develop less-heat for a given load, have higher by-design thermal "headroom",
and virtually all manufs have implemented effective motor-overload protection systems.
(won't stop the decoder shutting-down under extreme duress, but at least it will survive to fight another day).

Evidence of this evolution in decoder design, esp for High-current O and larger applications can be observed in the (now obsolete) ProtoCraft "Super Tsunami", which was a "you provide the donor TSU, we'll bolt on a beefier 3.5Amp Output Stage" upgrade service. With the release of the TSU2 4400 4Amp decoder, such high-current-capacity "kitbashes" were simply no-longer-required.

...Oh, and of course, it IS up to the modeller NOT to abuse the decoder with clapped-out, bound-up, magnetically-weak, brush-destroyed, short-circuited, electrically-noisy, agricultural motors which have no earthly-right powering any Operationally-Active locomotive in today's model RRing world...

...sure, if you're keeping that "special (brass?) model" in factory-stock "pristine" condition for "collectable" reasons, then by all means keep it in it's box, or a display case, where it belongs...

...if however it needs to earn-it-keep and pull-its-weight on an Operational DCC-powered layout,
then a proper motor-upgrade/replacement is in everyone and everything's best interests, esp the loco itself!

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Reply 0
greg ciurpita gregc

mosfet have very low resistance

the decoder mosfets in the h-bridge carry all the current to the motor.

with linear controllers, whatever voltage not across the load is across the transistor supplying that load.   the power is therefore the load current multiplied by the voltage across the transistors

but with PWM, the mosfet transistors are fully on or fully off, so there is minimal voltage drop across them and mosfets often have very little resistance, < 0.1 Ohm, so dissipate very little power even when passing larger currents.

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site

Reply 0
Reply