Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

AC/DC Transformer with 240V output

Simple safe test plan

one 9V cell .. no light
two 9V cells in series .. no light
three 9V cells in series .. light on means I am correct. but not ideal brightness

With an adjustable 1A DC Voltage regulator , you can adjust the brightness. These are small and inexpensive with a digital display.

My crystal ball says you are a photographer so can you send same quality photos of Led Lamp circuit board ?

That power supply you nicely photographed is usable only as I suggested and is otherwise junk.
 
My crystal ball says you are a photographer so can you send same quality photos of Led Lamp circuit board ?
You are correct. Just so I understand you want a photo of the LEDS in the circuit? I would have to cut the tube open to see the inside. Now I did connect it to this 9-36v regulator and no light even at its max 36V DC
DC regulator.png
 
The label on the device is obviously misprinted - surely you realised that?, how bright do you think LED lights would be fed with only 93uA :D

And post #2 still has the wrong current - so changed, but not corrected.
Yes, I realized that as I indicated in my original post, where I said: "There is something wrong about the numbers listed on the device:". LEDs operating at fractions of a milliamp makes no kind of sense. I guess I'm just not smart enough to figure out what the manufacturer meant, as if I care.
 
That Link in #26 is the exact same product
OK, There are better quality grow lamps but these are free. so ....

But faulty supply using two drivers means it will be extra bright with twice the power supplied and not last as long as expected.

If it is hot that's a clue there is a problem with the LEDs not matching the supply.


I use 48" Fluorescent T8 tubes that are tri-phosphor true daylight optimal for lighting but plants like UV Blue and Red in various amounts depending on plant, cycle and growth rate.
 

Attachments

  • Fluorescent light effects compare to natural sun.zip
    130.3 KB · Views: 59
BTW I get > 50kh lifespan on these 48" tubes
It turns out these tri-phosphor last longer if left on all the time,
over 10 yrs now = 87,600 hrs for kitchen counter light.
 
I use 48" Fluorescent T8 tubes that are tri-phosphor true daylight optimal for lighting but plants like UV Blue and Red in various amounts depending on plant, cycle and growth rate.
That's super that the bulbs are still going strong. I have a mobile home I rent and the original kitchen tube light from 1972 is still working.:D Thanks for the tip on tri-phosphor.
My wife took up gardening a few years ago. Im just getting her started with indoor growing. I knew the 5500k and the 91 CRI value were good enough too help the seeds & peak her interest. There is also a heat mat under the trays.:)
 
These tubes are rapid start without heaters using modern ballasts. Both end pins are shorted and not compatible with old ballasts. The Yellow wires are all common to all tubes on one end and they are no long dependent on all tubes installed and work independently.
 
View attachment 144652

Bad engrish translation.

This should output read 93 mA or 0.093 A and my reverse engineering on photo says it will be 100 mA.

The two bulk caps in series are rated for 250V so it can tolerate 500V * 80% (from tolerance mismatch)

No-load higher voltage is always higher when the max number of LEDs in the string limits the voltage.

I roughly estimate 3.3V per LED and think this can drive up to 54 series LEDs with constant current.

What does the customer datasheet say for the number of LEDs minimum in series?

This is just a simple 100 mA linear current limiter using 7.5 Ohms across 0.75 Vref meaning the IC will get hot as a low dropout linear regulator if the dropout is not "low" meaning if you do not have ~ 54 white/blue LEDs in series = 178V out, with a load of <= 16.5W, the IC will burn up. This is a product "worth about 25 cents". and consumer cost about $5 more or less is a ripoff.

It is meant to be around the ceiling perimeter of a home theater out of reach by hand for non-insulated electrical hazards and not safety approved.

Look for something worth about 25 cents per Watt with full engineering specs for minimum and maximum LED power current and voltage.

1) 110 V x 0.275 mA = 30 VA (vector or apparent volt-amps)≠ 18 watts. It is more like 30 mW
2) 178 V x 0.093 mA = 16.55 mW, which means the thing is pretty inefficient, just over 50%. That is piss poor in my estimation.
Looking back at measurements in #1 over 200V and 246V no load could actually be a calibration error using AC mode to measure DC (a common measurement error for meters).

Looking at the outside box, I recall parts of Japan have regions of 110V grid and I am wondering if this product was designed for that market only.

So it would seem to prevent burn out one the current limiter chip and one needs to drop the input voltage from 120 to 110 by some power resistor. 10V/0.1A = 100 Ohms @ 1W @ 150'C or a 5 W resistor in series with the very hot 50 Ohm 2W resistor at the AC input.

North America is generally 120V +/-10%. others may be 127 V nom. Mine is usually 120~123 Vrms. That could burn out the current limiter which becomes a variable resistor and heater with 1 watt for every 10V rise @ 0.1A Also, 1 Watt is all this chip can handle @ exceed 150'C and that's pushing it already. So it must have been designed for 110 Vac sources ONLY and not North America which is why they are unreliable.

This product just rectifies to DC with 4 diodes so 120 * sqrt(2) = 170 Vdc with no load so the measured No load in #1 is high 246V but this is a measurement error where meters will measure almost double the DC value in AC mode intended for AC only but when with DC are almost double. (std. rect. avg. to rms scaling).

I should not give any more advice in case you injure yourself measuring high voltages inside.

I recall a University in Florida and an engineer who designed grow LED engines there but the link is not handy.
 
Last edited:
I should not give any more advice in case you injure yourself measuring high voltages inside.
You are a wealth of information. I don't need to do any additional measuring. You and the rest of the members have explained what this is and how it works exceptionally well and after I am finished digesting all the info I'm gong to need a couple of 5 second micro-naps. :hilarious:
 
The 7 pin IC is an active controller of some sort - possibly one of these?

Yes something like this would make more sense with adapt to a wider range of AC input but not quite a match on pins or external components.

That would work much better with a very minimalist number of external parts using the large magnetic part as a flyback coil in buck mode instead of a low pass filter.

So I believe the unit in the photo is missing a diode and may be the faulty unit as it a screened glue spot with a missing component. Otherwise the diode is hidden in the photograph behind the large magnetic body.

The input interface may not match and does not use PWM or DIM but it's a close match on the output.
1708899497274.png


or similar to below
1708900403799.png
 
Last edited:
So I believe the unit in the photo is missing a diode and may be the faulty unit as it a screened glue spot with a missing component. Otherwise the diode is hidden in the photograph behind the large magnetic body.
Yes that is the faulty unit. I could add one. I don't have an M7 but I should have a 1N4007
 
Yes that is the faulty unit. I could add one. I don't have an M7 but I should have a 1N4007
The 1N4007 is a standard reverse recovery time diode, which is fine for 50/60 Hz line frequency rectification.

But it is far too slow for use in a high frequency switch mode power supply.
 
The 1N4007 is a standard reverse recovery time diode, which is fine for 50/60 Hz line frequency rectification.
All the diodes on the board (See post #1) were M7 I assumed the missing diode at D5 was also an m7. I will eventually open up another driver that is working correctly and see what the diode at D5 really is.
It says that the two are interchangeable. M7 & 1N4007.
 

Latest threads

New Articles From Microcontroller Tips

Back
Top