Don't want to divert any more of Beatsy's thread on the Godox AD200 as a macro strobe, so started another here for the Godox (Adorama) SK300II for the same use.
Earlier today I decided to completely disassembly the SK300II to get a better feel for the design concepts. Already knew the construction is quite good and the design seemed good as well. Originally I had posted that it looked like a DC to DC converter based concept rather than the usual voltage doubler.
Well, I was wrong and right!!
First off the energy storage is twelve 330uF 400V capacitors. Invoking the Energy storage is 1/2 CV^2 equation and solving for V, one finds V=389.25 volts. A classic voltage doubler, doubles the line peak voltage minus a couple diode drops, you get V=2(VLine*1.414)-2Vd, or V=337.4 volts with Vd~1v.
However, as mentioned, you need ~389 V for 300WS output. The SK300II does in fact invoke a voltage doubler design I discovered, so either Godox and Adorama specs are misleading, or something must be adding the extra "top off" voltage?
I believe the latter is taking place. There seems to be extra circuitry and large inductors (thus my original DC to DC converter speculation) that hints at the possibility of some type of inductive voltage boost concept (a variant of a DC to DC converter) taking place, possibly a regulated version of such with a TRIAC variable voltage doubler. Feedback regulates the capacitor voltage, either by varying the TRIAC timing, or the inductive voltage top off circuit, or maybe both.
Utilizing a higher energy storage voltage (389 vs. 338) may be what's behind the initial faster full power output pulse width (1/800 vs 1/220) since 300WS with 389 volts requires 3960uF while with 338 volts requires 5252uF. This produces a lower energy storage time constant with the effective resistance of the ionization tube.
Anyway, hope this helps inform folks looking into a AC mains powered Studio strobe for macro work.
Best,
Mike
Adorama (Godox) SK300II
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Mike, one way to check if the manufacturer is over claiming or not is to measure the terminal voltage of the caps. Make sure you set power to full when measuring it as some strobes use lower voltage for low power settings.
Be very careful when doing this, even though you know it.
If the large inductor is wired on the discharging side, instead of charging side, particularly if it is air cored, then it could be used to limit peak current for some reason(s) (some component used can not sustain large current, or to keep discharge curve well behaved, etc).
Be very careful when doing this, even though you know it.
If the large inductor is wired on the discharging side, instead of charging side, particularly if it is air cored, then it could be used to limit peak current for some reason(s) (some component used can not sustain large current, or to keep discharge curve well behaved, etc).
Peter,
Yes that would be a good conformation, assuming the capacitors values are accurate. I'm not sure what the tolerance is on these energy storage caps, I know they are special caps for this purpose. Since they are rated for 400v rather than the usual 350 volts, I'm inclined to believe that at peak output the voltage will indeed be near 389 volts, rather than the usual 338 volts with voltage doubler designs.
They have a beautiful common mode differential pi-filter on the AC inputs which uses one of the large inductors, which also points to a possible TRIAC driven voltage multiplier or some form of switch mode operation (keeps the EMI down).
The whole design and construction of this particular model (SK300II) is not indicative of a low cost strobe. The Adorama Flashpoint Rapid 600 (or Godox GT600 ~$600) isn't electrically built as well as this strobe, neither is the Flashpoint DG600 (~$200).
Best,
Mike
Yes that would be a good conformation, assuming the capacitors values are accurate. I'm not sure what the tolerance is on these energy storage caps, I know they are special caps for this purpose. Since they are rated for 400v rather than the usual 350 volts, I'm inclined to believe that at peak output the voltage will indeed be near 389 volts, rather than the usual 338 volts with voltage doubler designs.
They have a beautiful common mode differential pi-filter on the AC inputs which uses one of the large inductors, which also points to a possible TRIAC driven voltage multiplier or some form of switch mode operation (keeps the EMI down).
The whole design and construction of this particular model (SK300II) is not indicative of a low cost strobe. The Adorama Flashpoint Rapid 600 (or Godox GT600 ~$600) isn't electrically built as well as this strobe, neither is the Flashpoint DG600 (~$200).
Best,
Mike
Peter,
Here's a note from yesterday on another site & thread on the SK300II. Thought you might be interested.
BTW someday I'll get time to get a Windows PC and get your controller and software up and running.
Best,
Mike
I've done a little more investigation into the SK300II strobe design and want to update my original comment with what I've found.
"The first thing I checked was the energy storage capacitors, 12 at 330uF each rated at 400 volts. This gives (1/2 C*V^2) a potential of 316.8 Joules (WS), so the rating of 300 seems reasonable. The electronic design appears to be based upon a AC to DC conversion by way of full wave rectification, then DC to DC conversion, rather than the usual AC to DC voltage doubler in cheap strobes. Potentially this means a tighter control of the optical power with a regulated DC to DC conversion. The electronic components seem of good quality, at least from just looking at them."
I disassembled the entire SK300II strobe and traced out part of the circuitry that stores the charge and fires the ionization tube. Above I mentioned it looked as though the design was based upon a regulated DC to DC converter. After further circuit analysis and tracing I've found the design is based upon a regulated AC to DC voltage tripler. A TRIAC is used to control the AC line partial rectified sine wave "duty cycle" such that a shorter section of a rectified sine wave is used to control the peak voltage. This is placed in a feedback loop to maintain a given output voltage for the energy storage capacitors which supply the ionization energy. The "duty cycle" reference is varied to control the output flash tube level.
Just wanted to update this information. Anyway, it works pretty good as a studio strobe for macro work and probably other studio work as well.
Here's a note from yesterday on another site & thread on the SK300II. Thought you might be interested.
BTW someday I'll get time to get a Windows PC and get your controller and software up and running.
Best,
Mike
I've done a little more investigation into the SK300II strobe design and want to update my original comment with what I've found.
"The first thing I checked was the energy storage capacitors, 12 at 330uF each rated at 400 volts. This gives (1/2 C*V^2) a potential of 316.8 Joules (WS), so the rating of 300 seems reasonable. The electronic design appears to be based upon a AC to DC conversion by way of full wave rectification, then DC to DC conversion, rather than the usual AC to DC voltage doubler in cheap strobes. Potentially this means a tighter control of the optical power with a regulated DC to DC conversion. The electronic components seem of good quality, at least from just looking at them."
I disassembled the entire SK300II strobe and traced out part of the circuitry that stores the charge and fires the ionization tube. Above I mentioned it looked as though the design was based upon a regulated DC to DC converter. After further circuit analysis and tracing I've found the design is based upon a regulated AC to DC voltage tripler. A TRIAC is used to control the AC line partial rectified sine wave "duty cycle" such that a shorter section of a rectified sine wave is used to control the peak voltage. This is placed in a feedback loop to maintain a given output voltage for the energy storage capacitors which supply the ionization energy. The "duty cycle" reference is varied to control the output flash tube level.
Just wanted to update this information. Anyway, it works pretty good as a studio strobe for macro work and probably other studio work as well.
Mike,
OK, until the terminal voltage is measured, there are too many floating factors.
Yes, energy storage caps are special, they must have very low internal resistance. When tube discharges, current can be in hundreds of amps, for a strobe like this, probably in at least 200A. P=I*I*R where R is the internal resistance, so P = 40,000*R, assume R is 50 milli ohms, then P will be 200W!!
Sure, duration is short, say 5ms, so each time it is discharged, total energy dissipated in each cap is 200*5/1000 = 1J, imagine you shoot 200 images continuously over 15mins, that is 200J energy in heat in a congested enclosure.
This is why some cheap strobes are not suitable for stacking, maybe for portrait or product photography, light usage. So your recommendation of this SK300 II is valuable as you have used it without issues. Take a look at this cheap strobe shooting 120 images in 10 minutes
OK, until the terminal voltage is measured, there are too many floating factors.
Yes, energy storage caps are special, they must have very low internal resistance. When tube discharges, current can be in hundreds of amps, for a strobe like this, probably in at least 200A. P=I*I*R where R is the internal resistance, so P = 40,000*R, assume R is 50 milli ohms, then P will be 200W!!
Sure, duration is short, say 5ms, so each time it is discharged, total energy dissipated in each cap is 200*5/1000 = 1J, imagine you shoot 200 images continuously over 15mins, that is 200J energy in heat in a congested enclosure.
This is why some cheap strobes are not suitable for stacking, maybe for portrait or product photography, light usage. So your recommendation of this SK300 II is valuable as you have used it without issues. Take a look at this cheap strobe shooting 120 images in 10 minutes
Peter,
Reminds me of the cheap Neewer strobes, I had 3 explode. The 1st I got replaced, but later 2 more exploded. I replaced the exploded voltage doubler caps, then replaced all the other voltage doubler caps in each strobe just of safety.
Yes, low ESR is important for these energy storage and recharge caps!!
I also had some Interfit (EXT-400 I think) 400WS strobes, that served well, but I did have a cooling fan fail in one and another the fan is making a noise...so it's probably going to fail soon.
I eventually moved to the Adorama branded strobes and have had good luck so far.
Best,
Mike
Reminds me of the cheap Neewer strobes, I had 3 explode. The 1st I got replaced, but later 2 more exploded. I replaced the exploded voltage doubler caps, then replaced all the other voltage doubler caps in each strobe just of safety.
Yes, low ESR is important for these energy storage and recharge caps!!
I also had some Interfit (EXT-400 I think) 400WS strobes, that served well, but I did have a cooling fan fail in one and another the fan is making a noise...so it's probably going to fail soon.
I eventually moved to the Adorama branded strobes and have had good luck so far.
Best,
Mike
Yes, the first time it happened I was standing next to the strobe on a light stand. It sounded like a gun shot going off, scared the heck outta me!! It blew the innards of the capacitor all over the inside of the strobe. The second failure wasn't as loud, but still sounded like a gun shot. The 3rd was more of a "pop" sound.
I^2R heating and subsequent heat build up in the series voltage doubler caps was the problem.
Best,
Mike
I^2R heating and subsequent heat build up in the series voltage doubler caps was the problem.
Best,
Mike
Peter,
I went ahead and took the SK300II apart again and made some voltage measurements. The voltage across the energy storage capacitors seems to slowly vary over a few seconds with a slow decay and quick recovery (ramp style). This leads me to believe the controller is actually an impulse or hysteresis type rather than continual feedback loop. The variation is about 3 volts.
I did verify the voltage at 1:1 output as ~389.5 volts which is ~300WS with twelve 330uF capacitors, line voltage was 119.6VAC. I did some more reading of the capacitor voltage at different settings, this is what I calculated for energy:
1/1 300WS
1/2 172WS
1/4 104WS
1/8 64WS
1/16 42WS
I also did some in-between settings in 0.1 increments:
1/2 172WS
+0.1 179WS
+0.2 190WS
+0.3 201WS
+0.4 214WS
+0.5 226WS
1/8 64WS
+0.1 66WS
+0.2 70WS
+0.3 73WS
+0.4 77WS
+0.5 81WS
I don't know what the actual output output or optical pulse width looks like, but seems this strobe is 300WS and has the ability to be finely controlled with the 0.1 stop control. With the R2 RF controller the output is limited to 1/3 stop control, not 1/10 stop. Also note that the output is not very accurate at lower power settings, but repeatability might be good.
A little math shows that Delta Energy= Delta V(2E/V), so the Energy vibration at the high output due to the 3 volt variation on the caps is 4.6WS and at 1/16 output is ~1.8WS. So this seems to agree with the specs at the high output for a 2% variation.
BTW This is NOT for those that aren't familiar with high voltage devices and systems, the voltage levels in these strobes are lethal, even long after being unplugged from the AC power source.
Best,
Mike
I went ahead and took the SK300II apart again and made some voltage measurements. The voltage across the energy storage capacitors seems to slowly vary over a few seconds with a slow decay and quick recovery (ramp style). This leads me to believe the controller is actually an impulse or hysteresis type rather than continual feedback loop. The variation is about 3 volts.
I did verify the voltage at 1:1 output as ~389.5 volts which is ~300WS with twelve 330uF capacitors, line voltage was 119.6VAC. I did some more reading of the capacitor voltage at different settings, this is what I calculated for energy:
1/1 300WS
1/2 172WS
1/4 104WS
1/8 64WS
1/16 42WS
I also did some in-between settings in 0.1 increments:
1/2 172WS
+0.1 179WS
+0.2 190WS
+0.3 201WS
+0.4 214WS
+0.5 226WS
1/8 64WS
+0.1 66WS
+0.2 70WS
+0.3 73WS
+0.4 77WS
+0.5 81WS
I don't know what the actual output output or optical pulse width looks like, but seems this strobe is 300WS and has the ability to be finely controlled with the 0.1 stop control. With the R2 RF controller the output is limited to 1/3 stop control, not 1/10 stop. Also note that the output is not very accurate at lower power settings, but repeatability might be good.
A little math shows that Delta Energy= Delta V(2E/V), so the Energy vibration at the high output due to the 3 volt variation on the caps is 4.6WS and at 1/16 output is ~1.8WS. So this seems to agree with the specs at the high output for a 2% variation.
BTW This is NOT for those that aren't familiar with high voltage devices and systems, the voltage levels in these strobes are lethal, even long after being unplugged from the AC power source.
Best,
Mike