|
Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members. |
|
Thread Tools |
14th Jan 2013, 8:09 pm | #21 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 14,010
|
Re: 78 series regulator problem
The 78xx-series regulators can sometimes be a real pain to tame: a long time ago I designed a piece of kit which used a 7805, a 7815, a 7812 and a 7912 (this last being a -12V regulator) to power some digital control circuitry. Tested in the lab, the first few prototypes worked just fine. Off to production the design went - and when we delivered to the customer we had around 1 in 10 of the devices failing randomly.
Turned out that under certain power-up conditions the EPROM board presented what amounted to a negative resistance across the -12V rail causing the regulator to indulge in some crazy low-frequency oscillations (1 to 2Hz) and everything to lock-up. The answer was strangely elegant: I added a 470uF electrolytic across the offending rail, fed through a network of a 10-ohm resistor and an 1N5401 diode in parallel. The capacitor could charge through the diode from the regulator (which saw a low impedance to earth) and discharge in a controlled way into the load through the resistor when the load needed it. |
14th Jan 2013, 11:13 pm | #22 | |
Pentode
Join Date: Oct 2010
Location: London, UK.
Posts: 139
|
Re: 78 series regulator problem
Quote:
__________________
"Broken we can fix, cr*p is a design issue" |
|
15th Jan 2013, 12:50 am | #23 |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,536
|
Re: 78 series regulator problem
Sounds like a call for a few fish spine beads! Resistors (esp ww ones) can happily run a lot hotter than PCBs (esp paxolin) can take!
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
15th Jan 2013, 1:20 am | #24 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,970
|
Re: 78 series regulator problem
When homebrewing stuff I just apply the finger test. After doing basic calculations, I just touch the regulator or other potential heat source as the power comes up. If it isn't more than warm after a minute or so, I leave it for 10 minutes and give it a sniff. If there's no smell of hot electronics, I gingerly touch it again, and if it's still not too hot I assume it's OK.
I fully appreciate that a cavalier approach like this isn't suited to commercial production, though some bits of kit make you wonder I do usually use a current limiting resistor in the input to a regulator, which limits the worst case current and also acts as a fuse. |
15th Jan 2013, 10:24 am | #25 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,244
|
Re: 78 series regulator problem
Paul, I think that's fair enough, and basically what I do. I must admit that buying an infra-red thermometer was a bit of an indulgence, but it's proved to be very useful for all sorts of things - not just electronics (e.g. balancing radiators ).
Nothing wrong with the "squalid" input resistor - I do that myself when the current demand is known to be relatively constant. I normally have a reasonable electrolytic after the resistor as well - as well as reducing ripple to the input of the regulator, this helps to remove any HF noise on the raw DC (it's surprising how much can be present - of course, the error amplifier in the regulator is much less effective at higher frequencies). But, as has been noted, power resistors do run warm (300C!). It's all a bit of a balancing act. FWIW, I generally prefer the LM317/337 range - for the price of a couple of resistors, they have slightly better performance and you can keep fewer parts in stock. I rarely need to go for more "exotic" types as it's generally easier (and cheaper!) to engineer the circuit so that it's less dependent on the quality of the power supply. |
15th Jan 2013, 2:26 pm | #26 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,400
|
Re: 78 series regulator problem
No problems with resistor "sqalidity" here either- I think it's often a wise tactic to limit device dissipation and spikes and surges. Of course, in an ideal world, supply would be tailored for minimum headroom and maximum efficiency but, in practice, often a 5V rail is needed in a 12V environment or a 12V rail in a 24V situation, etc. Or, as in the OP, simply that a particular output voltage transformer is available and convenient. I've often seen input side resistors used in professional/commercial implementations, so no shame there.
|
15th Jan 2013, 5:45 pm | #27 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,195
|
Re: 78 series regulator problem
Hi Gents, if you are not averse to modern SMPS solutions the "Simple Switcher" series form Nat Semi is a low dissipation way to go. Devices are usually in TO 220 and design software is on the web. If your supply is on the low side they can be configured as either buck or boost converters.
Ed |