(https://i.postimg.cc/CZdWbnkc/rmtsw3.gif) (https://postimg.cc/CZdWbnkc)
Still searching my way through the jungle of my tube amp project. When I ordered my power supply transformator I settled for an additional 24 V secondary. Can't remember why. But there it is. And I have a handful of 24 V relays - I seem to have ordered them some day. Why 24 V, and not the more common 12 V? I don't know. There must have been something about the higher voltage that appealed to me at that time.
Well, here I am. And the overwhelmingly dominant part of relay driver circuits take a V+ of 12V as given.
Could you help me calculate R.G.'s circuit for 24 Volts? What I gleaned so far is that I should use BC550 instead of the 3904 (or similar) because it's rated for >50V max. And I should use 4004 diodes instead of 4148s.
Aside from the diode and Q1, the only thing you'd *have* to change would be the LED resistor.
What voltage are you using to switch this on? If that's a bit voltage, you might need to bump up the 4K7s a bit to keep the base current reasonable, and then adjust C1 by the same proportion to keep the RC time constant the same. But if you're switching from +5V, no need.
HTH,
Tom
Ah - I did not explain myself clear, I'm afraid. The relay's coil voltage is 24 VDC, that's also the voltage I'm planning on using as switching signal via a simple on/off switch - I hope the "Off" signal doesn't have to be negative voltage. That'd be a bummer.
Also I hope this circuit isn't a pulse kind of thing - that'd be a bigger bummer indeed. Do you think I could "pulse" it with 24V? Or will I kill it?
Wish I'd have bought 5V ones AND a 5 V secondary to fool around with a Attiny or such, but I always can do this in another project.
I hate relay circuits - there's so many and my head is spinning already.
Maybe I should take this one? (also in 12 V :( )
(https://i.postimg.cc/7JSccQzt/relays-f4-3a.gif) (https://postimg.cc/7JSccQzt)
I think I'm suffering from early onset dementia. There's some aspects of electronics I just can't seem to get my head around and remember them ...
(https://i.imgur.com/kCRvPcR.gif)
39K resistor stands for +5V on D1 cathode / R1 left leg junction (about 450μA Base current..)
(you may need to tweak its value for Q1 fully-on for coil greater current..)
Wow, thanks!
I just tried the second one (from Rod Elliott's general article about relays) and took a 4004, a 32V Zener and 10k for the resistors. Since my "laboratory power supply" only shoots out 18,5V it worked so far. Now to Antonis' schematic. So cool!
Just also read edited clarification above.. :icon_wink:
39k is strongly depended both on Q1 current gain and on relay's coil resistance..
Put this on the board and it works (with 18.5 V, mind). Will connect my USB Osci tomorrow and check out if I'm actively frying something. Already ordered 7824s ... and I have a 24 VAC wallbug somewhere and a load of 4007s to rectify this. WIll suffice for testing pruposes.
Thanks guys!
I love the click a relay makes. It's very pleasant. Some childhood memory? Dunno ... maybe there's a story behind this ...
http://www.mrscolumbo.com/uploads/Relais.png (http://www.mrscolumbo.com/uploads/Relais.png)
(http://www.mrscolumbo.com/uploads/Relais.png)
QuoteWhy 24 V, and not the more common 12 V? I don't know.
FWIW,
48V telecommunications
24V industrial
12V automotive (not really 12V)
24V is used to avoid voltage drops over long wire runs.
Just curious why the 2N3904 and 1N4148 would need to be replaced here?
2n3904 has ratings vcb 60v and vce 40v and I’ll assume the relay isn’t pulling 200mA current?
And 1N4148 is rated for 100V vrrm. (I’d use a bidirectional TVS, but that’s another story)
What am I missing?
You are missing nothing at all.. :icon_wink:
QuoteAnd 1N4148 is rated for 100V vrrm. (I'd use a bidirectional TVS, but that's another story)
What am I missing?
The 1N4148 to 1N4004/1N4007 change comes from Ron Elliot's site.
IMHO it's fine to use 1N4148's *provided* the relay coil current isn't too high.
Ideally the diode needs to handle the relay current. The fact the diode current is
a short pulse reduces stress and adds reliability anyway.
Quote from: Rob Strand on December 13, 2020, 04:08:46 AM
QuoteAnd 1N4148 is rated for 100V vrrm. (I'd use a bidirectional TVS, but that's another story)
What am I missing?
The 1N4148 to 1N4004/1N4007 change comes from Ron Elliot's site.
IMHO it's fine to use 1N4148's *provided* the relay coil current isn't too high.
Ideally the diode needs to handle the relay current. The fact the diode current is
a short pulse reduces stress and adds reliability anyway.
I confess it was a 50:50 chance which diode was taken - I have both in abundance, and I think the distance to the drawer with the 400x was shorter that that to the 4148s :)
QuoteI confess it was a 50:50 chance which diode was taken - I have both in abundance, and I think the distance to the drawer with the 400x was shorter that that to the 4148s
Far more logical than rolling a dice or flipping a coin ;D
Thank you - as always - for the valuable insights. I put the 24V matter aside for the moment, since I found a very nice 240 VAC/12 VDC module in a drawer (seems to be useful even in medicinical applications), AND I found a handful of 12V non-latching relays which give me the chance to apply a lot of schems as are (12 VDC) instead of having to calculate for the double voltage.
But I would love to beable to calculate the correct voltage to turn on my BC550 in this configuration. Just to learn something ...
Quote from: KarenColumbo on December 13, 2020, 07:51:58 AM
But I would love to beable to calculate the correct voltage to turn on my BC550 in this configuration. Just to learn something ...
Nothing special, indeed..
(just consider coil resistance as Collector load..)
You just have to take into account Collector current (coil voltage/resistance) and BC550 h
FE..
(say I
C = 30mA and min h
FE is 100..)
For BC550 to be fully-on (saturated), you need a Base current greater than I
C/h
FE (I
B>300mΑ)..
Ignoring V
BE, you have 24 - I
B X (R1 +R3) = 0 => I
B = 24 / 49k = 490μA..
(plenty current for saturating BC550 up to 49mA Collector current)
A more detailed scheme:
(https://i.imgur.com/aVaXs8Q.png)
Wow, thank you, Antonis, that really gets me started!!
Do you have some patience left for me? N0000b time!! :icon_redface: :icon_redface: :icon_redface: :icon_redface: :icon_redface:
If so: Why won't this work? I mean, obviously I can toggle those LEDs, but the relay won't act. I think I'm shorting something out with the PNP snippet in there.
What I intended: If the switch is closed, LED 1 lights up. If the switch is open, LED 2 should light up. [[▲The 3906 is really a BC559, forgot to change that in Eagle, sorry]]
(https://i.postimg.cc/K1Fg9G7g/LED-Toggle-NPN-PNP.png) (https://postimg.cc/K1Fg9G7g)
Can't get what's happening..
Disconnect R4 left leg from switch lug3 to see if Q1 Base current is enough for saturating Q1..
P.S.
What is 12V power supply carrent rating..??
edit:
Try the below configuration, to save some items.. :icon_wink:
(https://i.imgur.com/wuMcZ11.png)
Quote from: KarenColumbo on December 13, 2020, 02:05:23 PM....If the switch is closed, LED 1 lights up. If the switch is open, LED 2 should light up. ......
You do not need the PNP. You have not said why you would need a Zener. _I_ prefer to sock the transistor's Base. And pull-down instead of a mere Open. Try this:
(https://i.postimg.cc/sBbJpkHK/LEDToggle-NPN-no-PNP-42.gif) (https://postimg.cc/sBbJpkHK)
Beautiful, thank you. Logic isn't my thing ... well, this concludes that :)
@ Zener: I got this from Rod Elliott's article about relays. ( https://sound-au.com/articles/relays.htm (https://sound-au.com/articles/relays.htm) It speeds up the deactivation of the coil and thus avoiding undue ageing.
QuoteBeautiful, thank you. Logic isn't my thing ... well, this concludes that :)
@ Zener: I got this from Rod Elliott's article about relays. ( https://sound-au.com/articles/relays.htm It speeds up the deactivation of the coil and thus avoiding undue ageing.
The stuff Rod Elliot mentions about the turn off times is all true. Basically you have a trade-off: The higher the voltage you allow across the coil the quicker it turns off. However the more voltage you alllow across the coil the more stress on the switching transistor. The higher voltage can be achieve with a series resistance or a zener. With the series resistor method the voltage starts high and then immediately start to drop. The zener keeps the voltage high for longer so the coil discharges quicker for a given maximum voltage.
IMHO, if the contacts aren't pulling any current (like audio signals) there would be very little reduction in ageing with faster turn-off times. There's no arcing causing contact stress. When the coil is turned off the contact force slowly drops as the coil current reduces until the spring wins. If you are switching high currents the point were the force is lowered can promote arcing and high resistance joints at the contact and that would reduce the life of the contact.
So for audio signals speeding-up relay switch off is solving a problem that doesn't need to be solved - well not unless you *need* fast contact turn off. There's probably a turn-off time that sounds less glitchy.
Quote from: KarenColumbo on December 13, 2020, 02:05:23 PM
Wow, thank you, Antonis, that really gets me started!!
Do you have some patience left for me? N0000b time!! :icon_redface: :icon_redface: :icon_redface: :icon_redface: :icon_redface:
If so: Why won't this work? I mean, obviously I can toggle those LEDs, but the relay won't act. I think I'm shorting something out with the PNP snippet in there.
What I intended: If the switch is closed, LED 1 lights up. If the switch is open, LED 2 should light up. [[▲The 3906 is really a BC559, forgot to change that in Eagle, sorry]]
(https://i.postimg.cc/K1Fg9G7g/LED-Toggle-NPN-PNP.png) (https://postimg.cc/K1Fg9G7g)
have you considered this current path?
(https://i.postimg.cc/bGXSSnF4/LED-Toggle-NPN-PNP.png) (https://postimg.cc/bGXSSnF4)
it can be resolved by changing the switch contacts around. the middle lug goes to the transistors, the outer lugs go to ground and 12V,
cheers, Iain