I bought a bunch of these last summer on AliExress to pulse 7-segment LEDs.
Do GND and NEG share the same connection?
Attachment 9604
I think it is (Darrel once corrected me about sharing grounds as a reference), but I'd like to avoid frying these.
Printable View
I bought a bunch of these last summer on AliExress to pulse 7-segment LEDs.
Do GND and NEG share the same connection?
Attachment 9604
I think it is (Darrel once corrected me about sharing grounds as a reference), but I'd like to avoid frying these.
Do GND and NEG share the same connection?
what does your multimeter say ?
if they are galvanically isolated then they don't
find the data sheet and/or a schematic to be sure
or
at least a picture of the other side of the board so the possibility of an opto coupler can be established
Nope, not connected together. No BEEP on DMM.
I have no idea what "galvanically connected" means.
But I did find 2 of the circuits on Proto Supplies:
Attachment 9605
Attachment 9606
One connection is GND earth, the other is GND power...?
Proto doesn't have the FR-120N, but I have found a picture of all 3 circuits elsewhere. They seem identical on the top except for the MOSFET used (2 share the same opto-coupler).
Attachment 9607
EDIT: The PWM comes from a PIC. The PIC and this board both will run on the same 5V supply.
I've just very rarely dealt with a earth GND before.
galvanically isolated means there is no physical (metallic or otherwise conductive) connection between the terminals .Quote:
I have no idea what "galvanically connected" means.
the terminals are optically coupled therefore galvanically isolated. therefore the gnd and the neg do not need to be connected and probably should not be connected
I've found more data from the listing.
Attachment 9608
This is meant to power motors; hence the use of the flyback diode we're supposed to add on the power connection, and the priority to isolate both sides.
I'm going to be powering LEDs, no mechanical devices. I don't really care about isolation, since both sides are getting power from the same source.
And that's how I answer my own question (I think).
:)
The idea of using optocoupler is to have electrically separate circuits. If you connect the two reference points together as a common ground then what is the point of using an optocoupler? Connect directly the PWM to the MosFet and save few parts.
Ioannis
I was trying out Richard's idea of using pre-made circuits. It wasn't the same design as what he used (no boost), but I thought it would be a good exercise.Quote:
Originally Posted by Ioannis
https://www.picbasic.co.uk/forum/sho...110#post154110
This circuit isn't to power the project. I'm just trying to use 2 pots and PWM 5 volts to control the brightness of LED strings and the intensity of 7-segment LEDs (one pot controls the dash lights using 1 circuit, another the radio display using a 2nd circuit).
I'm trying to reproduce this effect:
https://i.imgur.com/epym9bt.png
I do not quite follow what you are after. But anyway, the purpose of the optoisolation is to isolate circuits that are in different voltage levels for either safety of the users or the safety and protection of the circuits.
If none of these are of concern at your application, then I see no point in using that module. Connect directly the PWM signal to the MosFet and control whatever load you have.
Ioannis
Using JLCPCB opens up other options for me. It gives me access to very low-cost components with very small footprints that can be integrated in my circuits.
These PWM circuits are just for ease of testing now. I'll be switching to a dual logic-level MOSFET to drive the console lights and radio display.
The hard part is finding a device that is available at Digikey/Mouser/Newark in through-hole footprint, as well as the desired SMD footprint at JLCPCB.
I'm just starting to look for a good L-V MOSFET now. Any suggestions?
Quote:
Originally Posted by Ioannis
This circuit was only to provide a fast and easy source of current without worrying about transistor/MOSFET/whatever selection.
If it became my final solution, great, But it was just a way to test PWM LED strips and 7-segment displays without worrying about over-loading a switching device.
I'm doing that now (looking for a suitable L-V MOSFET).
Semiconductors are not cheap on LCSC or JLCPCB. Maybe sometimes twice the price of Mouser or Digi-Key. But on passives, in any case these are cheap.
As for a MosFet, there are thousands available! I would not bother too much and select any one that is on stock and is in my budget.
For example:
250mA https://eu.mouser.com/ProductDetail/...x%2FNVUw%3D%3D
4A: https://eu.mouser.com/ProductDetail/...T0hBeH7w%3D%3D
Ioannis
I really like one of those dual MOSFETs; that's exactly what I need.Quote:
Originally Posted by Ioannis
But JLCPCB doesn't have the 4A model, and I'd really like the ability to drive unlimited LEDs as high as I want.
I have these on hand and JLCPCB has them both:
IRFR5410 P-channel 1A VgsTH = 2.0-4.0V
IRF730 N-channel 5.5A VgsTH = 2.0-4.5V
But their Gate thresholds are higher than what google says is a "low-voltage MOSFET" (under 2.0V threshold)...?
Forum gurus say the device will turn on, but there's some doubt if it will "remain ON".
The LCSC and JLCPCB do not have common stocks.
Datasheets sometimes are tricky and need attention to details.
What are your needs exactly? If you aim for a lot of Id current then you will need a good gate drive too.
Ioannis
I do not know what you mean by that. MosFets are not Thyristors or Triacs. A MosFet will be on if the drive is within specs and as long as there is drive. It will not remail ON if there is no drive.Quote:
Originally Posted by Demon
Ioannis
Quote:
Originally Posted by Ioannis
I need to PWM this 3 foot 5V LED strip rated 10 watts / meter:
https://www.aliexpress.com/item/1005003307581709.html
Amps = watts / volts
Max of 2A for a glareshield (like a car dashlight).
I also need to PWM 7-segment LEDs on a 2nd PWM circuit:
https://www.aliexpress.com/item/1005003119570725.html
According to google, each segment is typically rated up to maximum of 20mA.
Max of 160mA for entire digit (7 bars and 1 decimal point).
Max of 960mA for 6-digits.
My uneducated instinct tells me to look for MOSFETs that can "at least" take twice as much, so they can remain cooler.
Quote:
Originally Posted by Ioannis
Sorry, I didn't keep a link of what forum I read that from. One of the "gurus" was saying it's not enough to turn ON a MOSFET, you must ensure it "remains" ON.
I also thought ON meant ON.
OK. If your drive is not within specs, your MosFet may not turn ON and this means it will be a resistor instead of a closed switch with low resistance.
Now, choosing a MosFet with 20Amps does not necessarily means that it will stay cool. For this to happen you must select one with low Rds ON resistance. Many recent ones do exhibit also that low Rds.
Finally you want it to be SMD or TO-220 ?
Ioannis
Quote:
Originally Posted by Ioannis
I need it to be through-hole only to test my code, production will be SMD. At $20 a pop, I will probably only use 2 to test the daisy-chain logic and circuitry.
The need to drive the 7-segment LEDs has evaporated with my recent discover of the MAX7219.
I'm only driving a 2A LED strip of 3 feet now.
Found on Mouser two choices:
At 0,781 euros per piece IRFIZ24NPBF, a 55V/13A, 70mOhm, Vth at 1.8V https://gr.mouser.com/ProductDetail/...E5KLNbCg%3D%3D
At 2,23 euros per piece FDP070AN06A0, a 60V/15A, 7mOhm. Vth at 2V. https://gr.mouser.com/ProductDetail/...Y6Zv7sew%3D%3D
With the second the power dissipation at 2A will be 0,028 Watts !!!
With the first 280mW. Also pretty good I guess.
Note that the first is marked as LC (reaching end of life or something).
Ioannis
The power consumption of the second one is too good, isn't it?Quote:
Originally Posted by Ioannis
That is what you get for that price!
Ioannis