Re: [EVDL] Reduce EV: Charging load on the grid ... LEDs (update)
Robert Bruninga via EV wrote: UPDATE: For grins, I had the LEDs from a second bulb. It fully illuminated at the same 80 volts over 14 LEDs at 60 mA, but was 100% linear among the LEDs. Every one of the LEDs started dim and ramped up linearly and identical to each other just as one would expect in a series string. A retest of the original set still showed the totally erratic threshold effect of each of the LEDs flicking on and off until all were on at the same 80vdc and 60 ma. -still unexplained- Don't 'cha just love modern "quality control"? :-) Maybe they got a bag of cheap LEDs with some internal circuit, intended for car lights, blinkie toys, or as low battery detectors. -- Problems that go away by themselves will be back with friends. Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com ___ UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub http://lists.evdl.org/listinfo.cgi/ev-evdl.org Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
Re: [EVDL] Reduce EV: Charging load on the grid ... LEDs (update)
UPDATE: For grins, I had the LEDs from a second bulb. It fully illuminated at the same 80 volts over 14 LEDs at 60 mA, but was 100% linear among the LEDs. Every one of the LEDs started dim and ramped up linearly and identical to each other just as one would expect in a series string. A retest of the original set still showed the totally erratic threshold effect of each of the LEDs flicking on and off until all were on at the same 80vdc and 60 ma. -still unexplained- -Original Message- From: Robert Bruninga Sent: Friday, July 27, 2018 10:04 AM To: Electric Vehicle Discussion List Cc: Robert Bruninga Subject: RE: [EVDL] Reduce EV: Charging load on the grid ... LEDs > If you're running the LEDs on 120vac, the voltage drop needed is much larger. The 60W equivalent I just took apart had a full 14 LEDs in series but still used a PWM controller. It took 80 volts DC to get the LED's to their proper brilliance. So just a small resistor will do. This is just the metal disk with the 14 LEDs all in series, no other electronics. What was interesting was that 3 came on at around 15 volts, then more and more came on as the voltage was raised. And they shifted back and forth at various levels until they were all on. Since the bulb was rated at 9 Watts, and assuming 80% PWM efficiency, then I assumed the LED's were rated at 60 mA. And that took 80 volts. All the flashing back and forth as each began to illuminate surprised me. It was almost as if there was a threshold effect. As a new LED decided to illuminate, then the series string suffered due to the greater current and so all the other LED's adjusted. Some ablve and some below threshold. All being in series, it looks like there is a significant variability in their characteristics. But once I hit 80 volts the current above that was linear. But at 80 is also where it got to 60 mA and I did not go much higher. I was using a regulated variable DC supply with no series resistor (Supply was current limited at 120 mA). PS, this divides out to be about 5.7 volts per LED. Much higher than the 2.8 mentioned? Bob ___ UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub http://lists.evdl.org/listinfo.cgi/ev-evdl.org Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
Re: [EVDL] Reduce EV: Charging load on the grid ... LEDs
Again, it was the LED board only.. No other parts other than the 14 white LED's in series. -Original Message- From: EV On Behalf Of Lee Hart via EV Sent: Friday, July 27, 2018 1:43 PM To: Electric Vehicle Discussion List Cc: Lee Hart Subject: Re: [EVDL] Reduce EV: Charging load on the grid ... LEDs Robert Bruninga via EV wrote: > The 60W equivalent I just took apart had a full 14 LEDs in series but > still used a PWM controller. > > It took 80 volts DC to get the LED's to their proper brilliance. So > just a small resistor will do. Right. And a rectifier. :-) 80/120v means a simple series resistor would make it 67% effient. Not bad. 40v at 60ma is 2.4 watts of waste heat. > What was interesting was that 3 came on at around 15 volts, then more > and more came on as the voltage was raised. And they shifted back and > forth at various levels until they were all on. My guess is that there is more to the circuit than may appear. The flashing could be a side effect of the PWM circuit's "brownout" characteristics. It's also possible that there are other parts inside the LED package itself, too. Some are designed for applications like 5v indicators or 12v car lights, and have their own internal regulator or limiter. All these LED lights are made in China. They will use whatever part falls to hand this week, regardless of its specs or source. > PS, this divides out to be about 5.7 volts per LED. Much higher than > the 2.8v mentioned? Was that actually measured at the LED? Or just 80v at the PWM input, divided by 14 LEDs? It is likely that the PWM itself has a considerable voltage drop, so this won't properly estimate the LED voltage. It's also possible that each LED actually has more than one chip inside. The 2.8v I used in my example was for the red LEDs inside a cheap car tail light. That's actually rather high for red -- typical red LEDs are usually 1.6-1.8v. But they were driving these red LEDs at excessive currents; about 100ma, when the part's data sheet says 60ma absolute maximum. (This is unfortunately a common practice). White LED voltages are generally higher; in the 3.x volt range. -- Problems that go away by themselves will be back with friends. Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com ___ UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub http://lists.evdl.org/listinfo.cgi/ev-evdl.org Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA) ___ UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub http://lists.evdl.org/listinfo.cgi/ev-evdl.org Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
Re: [EVDL] Reduce EV: Charging load on the grid ... LEDs
Robert Bruninga via EV wrote: The 60W equivalent I just took apart had a full 14 LEDs in series but still used a PWM controller. It took 80 volts DC to get the LED's to their proper brilliance. So just a small resistor will do. Right. And a rectifier. :-) 80/120v means a simple series resistor would make it 67% effient. Not bad. 40v at 60ma is 2.4 watts of waste heat. What was interesting was that 3 came on at around 15 volts, then more and more came on as the voltage was raised. And they shifted back and forth at various levels until they were all on. My guess is that there is more to the circuit than may appear. The flashing could be a side effect of the PWM circuit's "brownout" characteristics. It's also possible that there are other parts inside the LED package itself, too. Some are designed for applications like 5v indicators or 12v car lights, and have their own internal regulator or limiter. All these LED lights are made in China. They will use whatever part falls to hand this week, regardless of its specs or source. PS, this divides out to be about 5.7 volts per LED. Much higher than the 2.8v mentioned? Was that actually measured at the LED? Or just 80v at the PWM input, divided by 14 LEDs? It is likely that the PWM itself has a considerable voltage drop, so this won't properly estimate the LED voltage. It's also possible that each LED actually has more than one chip inside. The 2.8v I used in my example was for the red LEDs inside a cheap car tail light. That's actually rather high for red -- typical red LEDs are usually 1.6-1.8v. But they were driving these red LEDs at excessive currents; about 100ma, when the part's data sheet says 60ma absolute maximum. (This is unfortunately a common practice). White LED voltages are generally higher; in the 3.x volt range. -- Problems that go away by themselves will be back with friends. Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com ___ UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub http://lists.evdl.org/listinfo.cgi/ev-evdl.org Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
Re: [EVDL] Reduce EV: Charging load on the grid ... LEDs
> If you're running the LEDs on 120vac, the voltage drop needed is much larger. The 60W equivalent I just took apart had a full 14 LEDs in series but still used a PWM controller. It took 80 volts DC to get the LED's to their proper brilliance. So just a small resistor will do. This is just the metal disk with the 14 LEDs all in series, no other electronics. What was interesting was that 3 came on at around 15 volts, then more and more came on as the voltage was raised. And they shifted back and forth at various levels until they were all on. Since the bulb was rated at 9 Watts, and assuming 80% PWM efficiency, then I assumed the LED's were rated at 60 mA. And that took 80 volts. All the flashing back and forth as each began to illuminate surprised me. It was almost as if there was a threshold effect. As a new LED decided to illuminate, then the series string suffered due to the greater current and so all the other LED's adjusted. Some ablve and some below threshold. All being in series, it looks like there is a significant variability in their characteristics. But once I hit 80 volts the current above that was linear. But at 80 is also where it got to 60 mA and I did not go much higher. I was using a regulated variable DC supply with no series resistor (Supply was current limited at 120 mA). PS, this divides out to be about 5.7 volts per LED. Much higher than the 2.8 mentioned? Bob ___ UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub http://lists.evdl.org/listinfo.cgi/ev-evdl.org Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
