Electronics > Beginners
Reducing 555 discharge current
<< < (3/5) > >>
Gyro:

--- Quote from: SeanB on September 14, 2018, 07:07:44 pm ---Having looked at the die photos you can see discharge and the output stage use a similar size NPN transistor on die, so the current is going to be the same for both, 300mA, as both are fed from the same current source. The transistor probably is not a saturated switch during the beginning of conduction, with the high pulse of current, so likely gets to around 500mA transiently.

--- End quote ---

Yes, I've done the same exercise with die photos, it's still weird though that no-one has ever quantified it in a datasheet, either in the parametric or Absolute Maximum sections.

The 555 has all sorts of non-timer uses (push button toggles etc) where it might be useful to have an open collector output rather than the totem pole. It would be nice to have more than an educated guess at it's continuous current rating. As you say, size wise, it ought to be about the same as the totem pole pulldown transistor as long as it stays in saturation.
T3sl4co1l:
Long duration, use CMOS (7555, LMC555...).  Or any of the logic timers (74HC123, CD4047, 4098...).  If you need high current output, consider a gate driver instead of the old crappy 555's output; TC4420 is quite a bit beefier (also faster, rail-to-rail and low shoot-thru).

Tim
viperidae:
If you want a large range of time intervals, why not use a counter chip clocked by the 555? You can then select different counts and frequencies to obtain the desired time.  You also mitigate the issue of leakage in your capacitor when using very large timing resistors.
TwinOak:
Hello everyone, thanks for all the input. Given that a few different topics around the 555 have been discussed but not so much the one I intended, my guess is that it's not such a common problem.

SeanB and Malaga, it's not the discharge current I'm worried about, it's the steady state idle current through the charge resistor that I want to reduce.

Tim, thanks for the alternatives, I'll check them out. Even though it's not long durations I need, just large adjustable spans. Look at the 74hc123 for instance, it requires a minimum charge resistor value of 10 or 2k depending on Vs, and max 1M. That won't even give me 3 full decades of adjustment, best case.

Viperidae, probably the best suggestion so far, especially if precision and long intervals are required. But perhaps a bit more cost than a few passives if not.


So, should we turn it the other way, are there any big downsides to the FET/diode suggestion in the op, can you fault it?


Best regards
Alexander
T3sl4co1l:

--- Quote from: TwinOak on September 17, 2018, 10:44:44 pm ---So, should we turn it the other way, are there any big downsides to the FET/diode suggestion in the op, can you fault it?

--- End quote ---

Regarding that, the leakage of the diode, and the input bias current of the 555's comparators, are still present, so you don't get any upward extension in terms of period.  The transistor is just gating a resistor, which doesn't make much sense when that's what the traditional circuit does (resistor 1 from +V to DISC, resistor 2 from DISC to capacitor).

You could add a JFET source follower, to eliminate* comparator bias current.  Likewise, buffering DISC with a low-leakage diode (certainly not a schottky!) or transistor will eliminate that source of error, allowing very large charging resistors (not quite gigs).

*Well, reduce by a few thousand times, say.  There is no elimination of course, but ~pA leakage is possible with these improvements.

Or just get a CMOS version, which addresses pretty much all of this on-chip -- note discharge leakage current, and THR/TRIG input bias:
http://www.ti.com/lit/ds/symlink/lmc555.pdf

Tim
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod