Nikon sb-900 - calibration appears off

Hey Guys, I seem to be getting strange output levels at the lower and higher ends of the adjustment range. 

Flash with no trigger gives readings as follows. (128=f2.8, 64=f4, 32=f5.6, 16=f8, 8=f11, 4=f16, 2=f22, 1=f45) (seems to me like 1/1 SHOULD give f32, but regardless…)
Flash with V6, profile set to manual gives similar readings to above, with maybe a half stop power increase from 1/8 upwards.
Flash with V6, SB-900 profile gives (Lo=f4, 128=f4, 64=f4, 32=f5.6, 16=f8, 8=f11, 4=f16, 2=f22, 1=f90!!)

So, with remote power adjustment available, Lo seems useless, 1/128 overpowers the flash by a stop, and 1/1 by 2 stops. The other levels seem accurate.

What's up with this? I'm guessing those massive 1/1 dumps are not healthy for my speedlight.

I took all these readings in a darkened room with a light meter, and took several readings at each data point. V6 firmware is 1.0.173.



  • Hello Chris,

    Thank you for your detailed experiments. We love to find out the reasons and resolve the problem. It is somehow strange in the result of your test. I would like to point out some findings.

    Firstly, flash with no trigger that you probably set the SB-900 in Manual mode. The flash was fired at each power level and got those number. What I concern is the power at 1/1. As you mentioned, it should be F32, not F45. The F value comes with two decimal places on the light meter. Usually a small measuring difference is acceptable, like -+0.2EV. Is it possible that the measurement looks huge difference, for example at F22 .09 , then to go to F45 .01,  but indeed it is only +/- 2 EV.

    Secondly, the V6 cannot increase the power output of the flash at the max power. The SB-900 with V6 is not possible to fire beyond its max power in TTL mode or 1/1 in manual mode. What you mentioned (1=F90) is abnormal. We are not sure why it happened in your test, but if you can give us more instruction, we will be able to replicate that problem.

    The V6 is not a booster to SB-700. It is designed to control the power output within the flash's capacity. Don't worry! it won't hurt your speedlight at all.

    Third, the SB-900 has to set the EV compensation to 0 in TTL mode while using with V6. If you set it at other value, the output may be inconsistent. The original design of the V6 doesn't handle this case, but we will improve it in the future.

    By the way, the latest firmware is V1.0.175. Please give it try if that can fix those problem.

    Thank you.

  • edited August 2014
    Hi there,

    I've done more comprehensive set of tests, with slightly different results from what I saw the other night, although still quite different from what I believe I should be seeing if everything is working as it should. Regarding the difference from the other night's results, All I can say to that is that I repeated the tests multiple times tonight, with similar numbers, so I would believe these readings rather than those in my first post.

    I have here 3 SB-900s - all on firmware 5.02, and 5 Cactus V6 units - all on firmware 1.0.173. I only have a Mac at home, but will update the triggers to 1.0.175 next week at work. Looking at the release notes, there only seems to be one small change from 1.0.173 to 1.0.175, which would appear to be unrelated to the issue that I am seeing. I'm taking measurements with a Minolta IV F light meter. Multiple readings taken with different combinations of V6 triggers and SB-900s, with similar results. I also confirmed that the SB-900s had 0 EV compensation set.

    Here is what I am seeing:

    Flash set to manual mode: (128=f2.8, 64=f4, 32=f5.6, 16=f8, 8=f11, 4=f16, 2=f22, 1=f16.5-45). Note the variations in the full power pops. I don't have a good explanation for this, but I did note that if I made a second pop as soon as the ready light lit up, it would tend to be at a lower power, whereas If I waited a few seconds longer, I would get a higher power flash. Seems that the speedlight will allow the flash to fire even if the capacitors are not fully charged - perhaps to improve the perceived recycle time? Also note that between the 3 SB-900s, the highest reading I managed to get at any power level was f45.

    Now, hooking up the V6 trigger, SB-900 set to TTL mode (0 EV comp), and I get the following readings: (Lo= f2.8 128=f2.8, 64=f2.8, 32=f4, 16=f5.6, 8=f8, 4=f11, 2=f16, 1=f45-f90).

    So what I am seeing is this: From 1/64 power, all the way to 1/2 power, I was getting a full stop BELOW what manual triggering would give me. The 1/1 pops give me between 1 and 2 stops ABOVE the maximum I am able to get with manual triggering, which I find rather worrying. Both the Lo setting and 1/64 give me an identical pop to 1/128. 1/128 seems to be the only setting where I am am seeing a reading consistent to a manual pop. 

    I see your comment that the V6 is not a booster, and can only control the output within the flash's capacity, however my light meter consistently tells me otherwise.

    It would be interesting if someone else with SB-900s and a light meter were able to run a similar test, and see whether or not their results conform to mine.

    I also have a couple of SB-700s, and will run a similar test one night soon when I get the opportunity.

  • edited August 2014

    Thank you for detailed explanation.

     Chris_Court said: I don't have a good explanation for this, but I did note that if I made a second pop as soon as the ready light lit up, it would tend to be at a lower power, whereas If I waited a few seconds longer, I would get a higher power flash.

    Do you know Nikon SB-900 allows to fire even if the capacitor is not fully charged. To identify whether the SB-900 is fully charged, check the ready-light. When it is solid red light, it is fully charged. When it is off, the SB-900 is till able to fire, but the power level is not as strong as the assigned power level. That may confuse the testing result.

    Chris_Court said: Also note that between the 3 SB-900s, the highest reading I managed to get at any power level was f45.

    Now, hooking up the V6 trigger, SB-900 set to TTL mode (0 EV comp), and I get the following readings: (Lo= f2.8 128=f2.8, 64=f2.8, 32=f4, 16=f5.6, 8=f8, 4=f11, 2=f16, 1=f45-f90).

    It is weird that it resulted in higher reading and Lo power with SB-900 and V6. The V6 should enchant the SB-900 to produce more power output. The Lo power has been tested on the SB-900 with V6, it works as expected here. By the way, we will run a test again to replicate the problem.

    For your info., our light meter is SEKONIC L-358 . If possible , you can try using another light meter to run this test again.

    Before you run test on SB-700s, we found that the 1/1 power in manual mode is 0.1-0.2 EV stronger than the max. power in TTL mode. We put a black card in front of the camera and fool the flash to fire and the max. output. The reason is that SB-700 has a pre-flash before main flash. That would consume energy in the capacitor. While the V6 controlling power output of SB-700 by TTL, it is expecting the power output will be -0.1-0.2 EV lesser.

    Any information that can lead us to the problem will be helpful.

    Thank you and look forward to hearing you soon!
  • Hi there, thanks for working through this with me. Hopefully we can get to the bottom of the issue.

    I repeated the same procedure again today using a Sekonic L-308S (I held it right next to the Minolta meter, and they read within .1 of a stop of each other). Also, firmware has been updated on all V6 units to to 1.0.175. Lastly, I double-checked that the ready light on the flash was on before making full power pops, which has given me more consistent readings at this power level.

    Short story… the result is essentially the same (Lo=f2.8, 128=f2.8, 64=f2.8, 32=f4, 16=f5.6, 8=f8, 4=f11, 2=f16, 1=f64-64.5)

    Next, I tested a SB-700. Here are the results (Lo=f2.0.2, 128=f2.8, 64=f4, 32=f5.6, 16=f8, 8=f11, 4=f16, 2=f22, 1=f45). Basically what I would expect to see, with the exception again of an overpowered 1/1.

    Seems we have 4 variables.
    1) Triggers/trigger setting
    2) Speedlights/settings
    3) Light meters
    4) Testing methodology

    I think we can eliminate #3, as I confirmed readings with a second meter, and I can't think of anything to improve #4, so if we can focus on any differences between the triggers and speed lights I have here, and those in your test setup, then maybe we can find an answer.



  • Hello Chris,

    When you preform the test, what is the distance between the light meter and the flash? I suggest to keep the flash at least 1 meter away from the light meter. The result on the SB-900 is quite expected as it should. The power output changes constantly on each power level as on the V6. Here is my setting,

    SB-900 in TTL  BL FP mode, zoom 200mm, no EV compensation. the result is,
    Lo = 2.0 6 , 1/128 = 2.8 3, 1/64 = 4.0 2 , 1/32 =5.6 2 , 1/16 = 8.0 3 , 1/8 = 11.0 4 , 1/4 = 16.0 4 , 1/2 = 22.0 4 , 32 = 32.0 4

    I couldn't replicate the your problem at Lo , 128 and 64 that power output remain unchanged. That is different than my result. It supposes to change the power output constantly. Our flash has been reset by using the rest function in the Menu of SB-900. If you have time, would you reset it and test again.

    I previously test on SB-700, my result is similar to yours, except the power output at 1/1. Our finding is whenever the power output is set to 1/1 on the V6 and if the adjusting wheel add more step. The display has no indication, but in fact it has confused the power output by +/- 1 EV. It happens in 1/4 to 1/1 occasionally. We had a discussion about this issue in another thread , Nikon Speedlight SB-700 won't fire brighter than about 1/2 power  . You may have a look for your reference.

    I would consider the testing environment and the flash setting. If you let me know how you set up, hopeful this will lead us to the problem.

    Looking forward to hearing you soon!
  • OK, I have found something that might point us to a solution. It seems that the SB-900 zoom setting may be at the root of our problem.

    I observe that, regardless of what the flash zoom is set to before turning on the V6, as soon as the V6 is switched on, the flash automatically zooms to 200mm. 

    In my previous testing, I had not paid attention to the zoom setting. When I set the zoom level to 200mm as you suggested, BEFORE attaching the flash to the V6, then suddenly everything seems to work as it should. Great news!

    Out of curiosity, I made some further experimentation and it seems that so long as the flash zoom is set to 70mm or above (before switching on the V6), then everything functions as it should. With zoom levels set at 17mm-50mm, we start to see some strange behaviour, including that which I've detailed above.

    I am very curious as to whether anyone else can duplicate these findings.

  • The Fresnel lens which is translucent plastic panel in from of the flash is one of the major factor to illumination patterns. When the zoom set to wide angle , the light distribution on the edge will be more uneven. On the other hand, At the Tele setting, the flash produces a highly concentrated beam.

    Each flash has its own characteristic. Maybe that is the reason why SB-700 and SB-900 acted differently.

    I hope that will explain for your issue.

  • Well it seems that we've found out which setting causes the issue, but that doesn't really explain WHY it's happening. Perhaps it is something that can be fixed in a future firmware update?

    Were you able to duplicate the issue at your end?


  • Do you mean the power output is out of expectation when the zoom set at 17mm to 50mm? I just tested on SB-900 at 17mm. The reading is similar to when it is set at 200m. The SB-900 with V6 works as expected.

    The V6 has nothing to do with the zoom function. I doubt that is not the reason for inconsistency caused by the V6. By the way, we love to fix the problem for any known issues , but we are not able to duplicate your issue. Maybe you can give me a step-by-step instruction to re-produce this issue for our further investigation.

  • edited September 2014
    In my brief testing, firmware 1.0.184 seems to have addressed some of the issues that I was having, however I am still getting some unexpected behaviour, seemingly related to the zoom number that the flash (sb-900) is set to before the V6 is switched on.

    I will do some further measurements and report back.

  • edited September 2014
    Thank you Chris.

    I'm glad you share you finding with us.

    Ray Chan

    Senior Product Specialist
    Harvest One Limited

  • p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px 'Helvetica Neue'; color: #454545}
    p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px 'Helvetica Neue'; color: #454545; min-height: 14.0px}

    Cactus V6 11.  I have 3 units, one TX mode and two in RX mode firing Nikon SB 700 and SB 900.  The units are set to TTL and are controlable.  Problem is that when measuring flash power with Seckonic L-308 S in studio, the meter fails to record accurate readings.  Flash used without Cactus and in manual mode provides accurate meter readings.  I can only conclude that when used with the Cactus, it fails to turn off pre flash thus giving spurious results

    Is there any way or suggestions for turning off the Nikon TTL pre flash via the Cactus units?

  • edited January 2018
    There's no way you can turn off the pre-flash, except if you set the flash mode to M.

    But by doing that you will not have remote power control, etc. 

    Also, the power output of the flash may not be 100% in line with the power levels seen on the V6 II, if you need the utmost accuracy, you will need to get Sekonic's L-858D. 

    Sekonic advised that this is the only model that works with pre-flash.

    Thank you.

    Antonio Lao
    Brand Manager

    To help us better help you, always state the exact firmware version installed on your Cactus device(s), such as: "1.1.013", "NIK.A.001", "v.103", or "A06".

    TTL or HSS not working on Cactus V6 II and V6 IIs? Be sure to check hot shoe connectivity by doing the <CAMERA INFO> check.

    Feel free to suggest an improvement or share product ideas. Contact us directly at  At Cactus, we listen. 
Sign In or Register to comment.