>>> PWM Frequency List <<<

[Ty_Bower]

I hope I didn't cause any confusion. My first picture is of the 10 Hz strobe. I didn't bother to post the slow strobe. The 'scope images don't portray an accurate image of the slow strobe. Remember the input to this 'scope is a PC sound card. It is useless for measuring DC voltage. The low frequency respose is starting to suffer even at 10 Hz. If you feed it a steady signal, the input coupling saturates and you get a zero reading. The slow strobe should appear as a square wave ~850ms wide, but instead appears as a tall positive peak that decays to zero, followed by a tall negative peak when the light turns off.

It's still a really neat piece of software. I've cobbled together a very crude photosensor out of the Dell's LEDs connected to the ends of a broken headphone cable. The 1/8" mini stereo plug fits right in the line in jack of my sound card. Just about anyone could probably duplicate my results and measure the PWM frequencies of their own lights. Later tonight I'm going to check my HDS and my Lumapower M3.

 

[David_Web]

Yes but the image indicates that the on-time is "noisy" as it's pretty much green rectangles. Have you ramped up the scanning frequency to the same as the full on to see if it looks the same?

My guess it that it's still noise from the converter and that it changes depending on how much it needs to boost the voltage. Then the PWM can come in and change brightness.

 

[selfbuilt]

The strobe software with a red LED wired to a dollar store computer microphone worked great - thanks for the heads-up Ty_Bower! :goodjob:

Setup was easy - I just ripped off the head of the microphone, and wired the two exposed bare copper leads to the two leads from the LED (from an old photon clone). Sensitivity is excellent with my Audigy sound card. And the software is very easy to use.

I was able to confirm all my previous results taken with the DMM/LED or DMM/solar cell. In all cases, the sound card/scope/LED results showed variances of less than 0.8% to my earlier recordings, which is pretty impressive.

But this method also allows me now to confirm PWM freq and duty cycle for lower intensities (e.g. 5% modes, etc.) and slower strobe modes (e.g. 1 Hz) that my DMM/solar cell method wouldn't pick up. So far, no surprises.

This is most clearly an excellent tool!

 

[Ty_Bower]

Sensitivity is excellent with my Audigy sound card...

Yes, it is very sensitive. I was getting nowhere trying to use an LED connected to my DVM. With the software, it was very easy to get a useful reading off just about any light. I've tested my HDS lights.

HDS Basic 42
Lowest setting (programmed to be as low as possible) - 506 Hz
"Secondary" setting (programmed two clicks down from default) - 1.9 kHz
"Primary" setting (at default) - 15.8 kHz
Maximum setting (42 lumens) - can't measure

HDS Ultimate 60
Lowest setting (set as low as possible) - 300 Hz
"Secondary" setting (re-programmed ?) - 1.2 kHz
"Primary" setting (set three or four clicks down from max) - can't measure
Maximum setting (set at 42 lumens) - can't measure

I'm impressed at how well the sound card can pick up a signal from the HDS, even on its lowest setting. The HDS is very dim on low. I'm also impressed at the high frequency the HDS light uses. I wish the Rexlights and Jetbeams had similarly high frequencies.

I also checked the PWM on my Lumapower M3. Although this thread suggests it uses PWM, I am unable to detect it. This agrees with what my eye sees. I'm very happy the Luma M3 doesn't use PWM.

 

[greg_in_canada]

Has anyone measured the Fenix P3D, or is this a non-PWM light?

Thanks - Greg

 

[Art Vandelay]

FlashlightReviews.com says the P2D has current regulation instead of PWM. My guess is the P3D will not have PWM either.

 

[HiltiHome]

My Jetbeam MKII (no X) measured 95,3 Hz

Strobe: 14,4 Hz

 

[selfbuilt]

Got my DX versions of the Jetbeam C-LE and MkIIX lights, and they are exactly the same as their JB comparables:

Dexlight X.1: 118-120 Hz, depending on mode, same strobe modes as MkIIX (i.e. 14/5/10 Hz)
Dexlight X.V: 73 Hz on primary (medium) and low, same strobe as C-LE (8 Hz)

Also received my Rexlight 2.0 from Kaidomain:

Rexlight: 94 Hz on medium and low, strobe mode 10 Hz

 

[Tronic]

I just received a few lights from emilion

Jetbeam JET-u AAA

med: 314Hz
low: 314 Hz
max: no PWM
strobe: 7.6Hz

Jetbeam C-LE v1.2

med: 315Hz
low: 316Hz
max: no PWM
strobe: 7.6Hz

 

[HiltiHome]

My Hyperion CE-R measured:

low: 312 Hz
med: 312 Hz
high: no PWM
strobe: 7,6Hz

No visible flickering on med, but still little nerves on low.

 

[HiltiHome]

My new WolfEyes Storm-X3-LED measured:

high: no PWM
med: 250Hz
low: 250Hz
Strobe: 20Hz

no visible flickering to my eyes, but noticeable when pointing to a fan

 

[Phaethon]

PWM efficiency

Excuse me if i am hi-jacking this thread but it's the one with the most recent posts.
I have been thinking on the pwm in-efficiency claims and i understand that they are based on simple math, calculating the lux output based on the average Ah with which the leds are driver. However, i see a fault in this reasoning, as the theory reaches a conclusion based on a continuous drive model -which derives by averaging the feed current- and a current feed TO lumens output. My point is, that we should be focused on the outcome of pwm driving as it is perceived by the human eye, not a lux meter. Therefore, i believe that at a frequency of 100Hz, the eye can only see continuous light that consists of full intensity bursts, which have -let's assume - 700 mah feed. Compared to a constant driving of 1000 mah, the light output perceived by the eye should be the same, because the eye can't average the light emitted in actual time but only in relevant time. A light meter might measure the output in less lumens, as it has the ability to take infinite measurement during the 100hz cycles. The human eye cannot and therefore the light emission seems identical. The practical outcome is that we have the same light with significally less heat production and battery drain. Any thoughts?

 

[selfbuilt]

Hi Phaethon ... :welcome:

What you suggest may be true, but most of the discussion around PWM efficiency at CPF (in the non-technical forums, at any rate) focus on its advantages/disadvantages relative to constant current control for producing visually dimmer light. I think everyone is in agreement over the advantage of some sort of circuit-controlled dimming (as opposed to an inefficient resistor setup, for example).

To summarize the general view, current controlled circuits have the advantage of longer runtime for lower output levels (due to the fact LEDs are more efficient at lower current drive levels), and lack of visual strobing effect. PWM has the main advantage of simpler and smaller circuits (and cheaper?) with no tint shift due to the consistent current drive level.

PWM typically gets a bad rap here since current-controlled lights win out over PWM in terms of runtime at lower modes. But some of the newer PWM lights with >1kHz show no visible strobing, removing that general complaint. :thumbsup:

It's a pity EngrPaul is no longer around to update this thread. His input was always very valuable.

 

[Art Vandelay]

It's a pity EngrPaul is no longer around to update this thread. His input was always very valuable.

No longer around?

I have not heard anything. Did something happen?

It's only been a month since his last post.

 

[selfbuilt]

I have not heard anything. Did something happen?

He posted a goodbye back in early-mid summer. Since then, he's only posted on the site half a dozen times or so. You'll notice all the pictures from his threads are gone too. I don't know why he left, but I do miss his input in the reviews and modding forums.

 

[Phaethon]

Hi Phaethon ... :welcome:

PWM typically gets a bad rap here since current-controlled lights win out over PWM in terms of runtime at lower modes. But some of the newer PWM lights with >1kHz show no visible strobing, removing that general complaint.

Thanks for your prompt reply. The reason for my post is that i am finishing a project torch (literally :devil based on the edison edipower 20W led. I had to source a 10-15 degree lens and have a housing (edison provides NO housing or lenses for their top-power led???) from solid aluminium machined, for the setup to obtain sufficient heat dissipation. For added piece of mind, i am using a pwm circuit (based on a 555 chip) to drive the 17.5V, 1 Amp led in order to reduce heat production. Tests are very encouraging and i have no significant heat buildup with great output. Please note that the edipower 20W has a nominal output of at least 700 lumens at 17.5 volts. I look forward to finishing the tests and post duration specs and photos in a separate thread.

 

[elho]

Interesting old thread.

Using a photodiode in photoconductive mode and a DMM I got the following results:

Jetbeam Jet-µ: 73 Hz [Strobe is 10 Hz]
Wolf-Eyes Digital Cree P4 HO 3.7v-6.0v D26 4 mode: 495 Hz [Strobe is 13 Hz]
Wolf-Eyes Digital Cree R2 WC low Vf HO 3.7v-6.0v D26 4 mode: 530 Hz [Strobe is 14 Hz]

These results go along with my observations of easily noticeable low frequency PWM on the Jet-µ and unnoticeable PWM on the Wolf-Eyes drop-ins.
The results for the Jet-µ however do not go along with those already in the list.
My Jet-µ neither has memory nor beacon and looks different than the one in selfbuilt's review, so being different versions would be an explanation. Or the DMM is fooled by the constant component in this setup or something.

I guess, I should dig out some red LED... :thinking:

 

[EngrPaul]

It's a pity EngrPaul is no longer around to update this thread. His input was always very valuable.

I apologize for being absent for quite a while. I don't think I could possibly catch up with a comprehensive list. Too many new lights while I was gone! :mecry:

 

[elho]

I apologize for being absent for quite a while. I don't think I could possibly catch up with a comprehensive list. Too many new lights while I was gone! :mecry:

Well, you don't need to do all the work - I'd guess if you just update the list in the first post with the data from the later posts in this thread, people will notice and be motivated to measure and submit data about lights missing in the list just like I did.

 

[Dissipator556]

Would be interesting to know the strobe rates of the Gladius and Xiphos lights (they look identical to me in strobe rate), as well as my Nitecore Defender Infinity (which looks much slower than the Gladius rate).


If I had the time to rig something up myself to measure them, I would make it a DIY project. Not enough time lately- anybody else know the strobe rates off hand?