Seems like there's a small (but fatal) error: R4 & R5 can't be 0 ohm unless you want to short the supply. You'd have to raise them to some value, say 240 ohm for 20mA. In the off state, that 20mA will pass through 480 ohm, generating
Apart from the resistor values, @matt_garman is correct because R4 and R5 keep the current loop closed.
Unless I'm not reading that schematic correctly, I believe the relays in that schematic are non-latching. There are two types of latching relays (at least in my application), single-coil and double-coil.
The single-coil non-latching and two-coil latching are "easy" to implement with the single transistor, single flyback diode protection scheme (per coil). These relay types are designed to have current flow in only one direction through the coil(s).
But I don't think a single transistor and single protection diode scheme can be used for the single-coil latching relay type. This relay is designed to have current flow in either direction, depending on whether the relay is to be set or reset.
Can you show how that would be used with the single coil latching relay type?
it is a nice part but more expensive, your money, your choice.
and yes the schematic is for non-latching relay, for latching you modify it.
Expensive? 5 for £3 on Ebay?
I thought from first post you were already driving relay direct from mcu pins and were looking for a buffer.
Your schematic in post #8 - replace the resistors with more transistors and you have a H bridge which is the usual way of reversing current: It's a transistorised DPDT switch. Also, use protection diodes.
Heed the advice in relay data sheet to spec 4.5V coil for 5V transistor switching.
My bad.
I thought from first post you were already driving relay direct from mcu pins and were looking for a buffer.
Your schematic in post #8 - replace the resistors with more transistors and you have a H bridge which is the usual way of reversing current: It's a transistorised DPDT switch. Also, use protection diodes.
Heed the advice in relay data sheet to spec 4.5V coil for 5V transistor switching.
russc said:
Yes, that is true. And it appears to be a somewhat widespread practice (at least in the guitar effect pedal world). But per this discussion (and others I linked in my original post), it's clearly not a robust design.
But my goal was to try and figure out exactly what that ideal/robust design looks like.
russc said:
Yup, that's more or less what I've figured. It looks like, for latching relays: single-coil requires four transistors; but a double-coil only requires two transistors. In short, it appears that the double-coil latching relay has the simpler implementation.
I was just trying to see if I was overlooking something, and if in fact there is a simple way to (robustly) drive a single-coil latching relay from an MCU ("simple" meaning same number of components or fewer compared to the double-coil latching relay).
For completeness, I will add that another simple way I've seen to drive the single-coil latching relay is with an electrolytic capacitor in series with the coil. Bi-directional current pulses are possible from capacitor charging/discharging. See here for an example.
Sounds more trouble than it's worth.
Shouldn't be surprised if 4 transistors are cheaper than 2 transistors + cap.
More info here:
https://www.te.com/commerce/Documen...glishENG_DS_1-1773931-5_0717.pdf1-1773931-5
Shouldn't be surprised if 4 transistors are cheaper than 2 transistors + cap.
More info here:
https://www.te.com/commerce/Documen...glishENG_DS_1-1773931-5_0717.pdf1-1773931-5
and an ic solution
https://www.eetimes.com/ic-drives-up-to-four-single-coil-latching-relays/
https://www.eetimes.com/ic-drives-up-to-four-single-coil-latching-relays/
Interesting ic, but not a hafbridge/fullbridge output, so relay will get less voltage through resistor ,and relay must be more sensitive (higher ohms,less current,more expensive) . Otherwise , resistor must be lower ohms and this increases load on ic outputs. But because all this is just short pulse , usable solution.
I used different approach, for input switcher in amplifier . Simple button , two coils latching relay, two resistors, two capacitors. Trick was to select capacitors capacity , for relay switch just once for single button press. One contact group switched which coil will operate at next button press, other group switched the signal relay coil. Used a 5V latching dual coil relay and 12V power supply, with 5V this was not possible. I will post schematic ,if i will find it in my papers from those old times.
I used different approach, for input switcher in amplifier . Simple button , two coils latching relay, two resistors, two capacitors. Trick was to select capacitors capacity , for relay switch just once for single button press. One contact group switched which coil will operate at next button press, other group switched the signal relay coil. Used a 5V latching dual coil relay and 12V power supply, with 5V this was not possible. I will post schematic ,if i will find it in my papers from those old times.