This is a placeholder topic for “Gravity: Magnetic Latching Relay for ESP32 / Arduino” comments.
This DFRobot magnetic latching relay is a new type of relay. It remains in the last position it was when last powered and features noiseless and low power…
Read moreIs this the only latching relay in Core Electronics catalogue ? Latching seems highly desirable, especially when the low power requirement is added in. Maybe it is also known by some other term ?
A google search revealed a large number of latching relays at other electronics stores - but all obviously designed for automotive or industrial use - not presented in a breakout board format convenient for hobbyists.
Hi Donald
I had a bit to say regarding this item in a previous post of yours so I would refer you to that.
In addition regarding pulse width. The text says 2mSec, The product Wiki says 4mSec and the data sheet says 100mSec.
Edit: And the Arduino sample sketch indicates 2mSec.
Please refer data sheet here
The relay pictured shows a 12V coil and I don’t see any external coil connection. The text says 3.3V or 5V operation. The relay (according to data sheet) seems to be a single coil latching type. There is a chip on board which might take care of the voltage reversal required to change state (H bridge???) and pulse stretching but as there is no schematic that I can find or any other information regarding the operation all that would be on a suck and see basis. A highly undesirable situation in my book.
The choice is yours
Cheers Bob
2mS is the module pulse time - this will be the minimum pulse width required by the controller chip, so it should match the code example. The product Wiki says the relay set minimum is 8mS and reset is 5mS: that will determine the maximum repetition rate. The 100mS you mention is the relay pulse time and is not related to either the relay set or reset time or the module pulse length.
Hi Jeff
The data sheet also says 8mSec and 5mSec for set and reset times
This I would suggest would be the actual relay operate time which is pretty fair for these small style relays
This says "Pulse Width min.
Which to me means just that. Pulse width min. I would suggest that that pulse width is to allow the relay armature to magnetise fully and hold the relay in its chosen state.
As I suggested this is probably taken care of with the smarts on that board. The above is taken from the actual relay data sheet.
This is the other discrepancy I was getting at. The product Wiki says a pulse width of 2mSec in most places but in the “notes” bit we find
quote
the excitation voltage applied to the coil must reach the rated voltage, and the pulse width must be greater than 4ms.
unquote
Leave one a bit up in the air when deciding what is happening.
Cheers Bob
and I am NOT going to purchase one just to find out.
Hey @Donald23173,
We do offer some other latching relays/relay boards, albeit the section is pretty small. See below:
Adafruit Featherwing Latching Mini Relay
Waveshare Modbus RTU 8-Channel Latching Relay Module
I am unsure if these fit your requirements, but worth looking into if you are not sold on the DFRobot variant.
You did. I did read and took on board. Which is why i asked this question.
Thanks Zach. I had in fact found both those. I’m only after a single relay so the Waveshare seems an overkill; and the Adafruit … well honestly I was overwhelmed by all the header pins. I expect that if I was familiar with Adafruit Feather it would all make obvious sense. Given the advantages, I was surprised to find so few latching relays in the hobbyist/maker segment of the market … and wondered if I was missing something.
I think the DFRobot will be ok. My requirement is for the relay to stay switched when my ESP32 goes into deep_sleep (ie is powered off); and since i’m not going to be switching fast, increasing the pulse time shouldn’t be a problem.
Hi @Donald23173
This relay module should be fine for what you’re looking to do.
Hi Dan
Please note the following, copied from the relay Data sheet
This would indicate the relay has a 12V coil.
The Fritzing diagram in the product Wiki shows the board connected to Arduino 5V. How does that work.
Question: What is the relay fitted to the boards Core have in stock.
Cheers Bob
Hey @Robert93820,
Unfortunately, we don’t currently have any stock of this item, so I can’t check empirically at this time. I have, however, done a bit of investigative work on the device.
There are two IC’s present on the device, an SSP8023D, and an SDB628
SSP8023D Datasheet
SDB628 Datasheet
The SSP8023 is a Bi-Directional Relay Driver with an input threshold of 1.5-2V, from the looks of it, it seems to be actuated with a logic-level input and output a voltage to the relay.
The other device, the SDB628, seems to be a current-mode step-up converter. The surrounding circuit on the PCB seems to comply with the “Typical Application” circuit seen on the datasheet.
In short, it looks like the SDB628 steps-up the device supply voltage to 12V, this is then used in conjunction with the logic inputs from the control device, and a Bi-Direction Relay Driver.
Hi Zac.
Thanks for that. Explains a lot. The photo provided does not make it easy to determine just what the 2 chips are. It still does not explain the discrepancy between 2mSec and the relay requirement of 100mSec. The data sheet for the SSP8023D also mentions 100mSec and although not explicit it could read that this chip outputs a 100mSec pulse to drive the relay. Something like this has to happen as it requires somewhat longer than 2mSec for the relay to operate then you have to have time for the magnetisation to be complete to hold it.
It also does not alter the fact that had this information been more readily available it could have saved a fair bit of to-ing and fro-ing and discussion in this thread alone. Unfortunately this sort of thing is getting very frequent. As if the manufacturer are frightened someone will rush off and build their own device. At the price some of these things are offered who would be bothered.
I do however insist on being able to easily evaluate a device’s fit for purpose status which in some cases (I admit not all) is nearly impossible. In these cases I would not consider purchase or comment on this Forum re suitability. I would go elsewhere.
As it turns out this device will probably suit Donald nicely. But it would have been nice to say that with some certainty a few days ago wouldn’t it.
Cheers Bob
Jumping in here to summarise for @Donald23173.
Yes! This relay module should be suitable for your purpose to switch and hold a state when your ESP32 goes into deep sleep.
While the relay itself has a 12V coil, the other components on the board will step up the logic level voltage used as input to that voltage so the relay works as expected.
While I agree with @Robert93820 that the differences in reported pulse width are frustrating, it doesn’t seem like this will affect your particular use case. If this does prove to be an issue for you in the future please let us know and we would be more than happy to lend a hand.
Once we get some stock in of this relay we can run a few tests on it to answer some of the questions raised in this thread with a higher degree of confidence.
Hope this helps!
Hi Samuel
Yes I found that out courtesy of Zac’s reply.
Quite right there.
Both of these items are very disturbing. By not saying anything the text (Core seem to be a verbatim copy of DFR which in itself is a worry, no checking) is pretty confusing. DFR in the “Specs???” quote 2mSec yet in the text body somewhere say a minimum of 4mSec and the Data sheet for the relay says 100mSec. Nothing on the voltage conversion to 12V and no schematic to help find out. The chip types are unreadable in the provided photo.
Could one of your tests be to apply a 2mSec pulse and measure the actual pulse width applied to the relay. I would not be surprised if it was 100mSec. 2mSec is not enough time for the relay to actually operate.
Cheers Bob
In case you miss it and have some sort of catastrophic accident the pulse to the relay is floating, it has to be reversible. So no ground connection. for this reason you will not be able to measure both pulses with the same scope. Even using 2 instruments you will have to remove the mains earth from the one measuring the relay pulse and keep the chassis isolated. Or use a battery operated tablet scope for this measurement.
And you can get our latest projects and tips straight away by following us on:
