still doing it with a 1N4007 flyback diode installed.
could it be the ESP noting the restore mode?
restore_mode: RESTORE_DEFAULT_OFF
Hi Anthony
A 1N4007 is not going to do much. The back EMF flyback voltage spike will be pretty fast and I doubt the 1N4007 would be fast enough to catch it. To be sure you need a fast schottky in that application. As far as high frequencies are concerned the 1N4007 might just as well be a piece of wire. I know a lot of people use the 1N400X series across relays but they don’t catch the initial fast bit of the spike (as observed many times on an oscilloscope). Work OK for a while but problems usually arise down the track a bit.
As for your copper overcoat I very much doubt you are getting any meaningful radiation interference from a relay coil but as long as you are not shorting anything it is doing no harm. For magnetic radiation steel is better anyway. I know the braid is copper on Co-ax cable but that is for convenience and flexibility. Not real good anyway, for something like 80% double screening is used. Where serious screening is required this braid is solid copper or for real serious screening a steel pipe or duct.
Cheers Bob
1 Like
Bob,
i have some other flyback diodes, purchased a pack. I have a 1N5822 as you suggested, but its a big hunk of as thing and I chickened out soldering it in. photo below. not even completely sure i put it in correctly TBH.
The copper tape is definitely doing something. I bench tested the device with and without and it didn’t drop connection with the API server. Just the following log entry:
09:10:40][D][switch:016]: 'tank switch on off' Turning OFF.
[09:10:40][D][switch:055]: 'tank switch on off': Sending state OFF
[09:10:54][D][esp32.preferences:114]: Saving 1 preferences to flash...
[09:10:54][D][esp32.preferences:143]: Saving 1 preferences to flash: 1 cached, 0 written, 0 failed
I believe the log entry is benign. I have a restore_mode set on a switch.
Hi Anthony
No Photo.
They make diodes a lot bigger than that. But yes, it is twice the size of a 1N4007 and LOTS bigger than the 1N4148. I think the leads are a bit thicker too.
If it is not soldered it may or may not be making a connection.
Probably is but I am not convinced that interference from the relay is your problem. There is a magnetic field sure (after all that is how they work) but it would not be all that strong. Have you anything else that is overly sensitive to magnetic fields. What I am getting at is people including myself have been using relays in the vicinity of other sensitive electronic devices for many years without undue interference.
When the relay operates so does the solenoid. Have you disconnected that to try. That would have a far stronger magnetic field than the relay and probably require more current. Don’t have the solenoid details so don’t know. If that solenoid is operating a water valve it could be quite strong and in proportion requiring a fair bit of current, more than the 12V supply could manage and could be causing a momentary glitch. I would even try using a larger 12V battery to try if you can get hold of one.
As I stated above the power source is the heart of everything. If it is not up to the task nothing will work. If this bit is in doubt you will have to remove that doubt before anything else or you will find yourself going in circles nd be like the cat swallowing its tail.
Cheers Bb
1 Like
Rob, i’ll see how it goes for the moment, it’s deployed and working in a field test. I can always change the diode later.
The photo didn’t upload. Its a forum permissions thing.
By ‘not soldered in properly’ i was meaning ‘placement and direction’ have the ‘band’ facing the positive wire, its soldered in well. its close to the gravity card.
Hi Anthony
That is correct. If it wasn’t you would have shorted out the relay coil and probably destroyed the switching Mosfet.
Cheers Bob
