Vivitar 285HV Modification - Replacement Filter Options

[Andrew Dayer]

In my efforts to get to the first base of UV photography, I brought a few Vivitar 285's and started converting them according to guidance here etc - including precautions for risk from the dangerous high voltage circuitry.

 

So far I'm not too unhappy - two successfully 'converted'; one more dead (seems to have been a late model with incredible fragile soldering; now charges but won't discharge).

I noticed, however, that the UV output of when using one flash is quite low and the unit has to be very close to the subject at full power to get half-decent exposure. This is limiting my aperture and sensor ISO choices. Ideally, I'd like to use the same flash position for visible and UV with the same camera setting (just changing the filters) but I can't reduce the output enough (min 1/16) for visible light so I'm having to use a second unconverted flash further back for visible comparison.

 

For the conversion, I've used a replacement window over the flash tube and reflector made from CD case plastic. I think this is likely to be a polystyrene-type material. It does appear to transmit a good amount of UV at 365nm based on a crude torch test (with and without subject material, UV torch and visible fluorescence of 'white' items with optical brighteners).

So, is the low output due to a coated flash tube, inherently low UV output, absorption by the CD material, or absorption by the zoom head fresnel (OK with UV torch)

Would using UV transmitting acrylic, UV glass, or fused silica (quartz) likely work better? I'm guessing no window is not recommended?

 

In addition, is there any info on how far into UV I can expect these units go? It seems to depend on the grade of quartz used to form the tube but I don't seem anything specific to these.

For fluorescence work, I presume that windows made from ZWB2 would suffice?

 

Seems like a lot of questions; any advice gratefully received!

[Andy Perrin]

Plastic does not transmit UV well. The test you made was very crude also. Even 405nm visible light makes a lot of fluorescence. Getting rid of plastic elements seems like the first thing that should be done. Ulf has also mentioned that the proportion of UV of most flashes is just not that big, so you need a fair amount of power anyhow. 

[Doug A]

I modified a Vivitar 283 for UV. I didn't even bother putting any window over the flash tube. It still requires ISO 800 and up to shoot macros at F11. This is with the flash ~6" away from subject. I was hoping the Vivitar would output a lot more than my modified Pentax 540 flash. Not happening. I usually use both flash together. 

 

@Andrea B.uses a shoe mount Nikon and fires it 3 times. Multi popping is great as long as there's no movement. @Adrianuses 4 modified Metz handle mount flashes and @nfotoemploys giant studio flash units. Getting enough UV light is challenging. 

 

Thanks,

Doug A

[Andy Perrin]

Doug if that thing explodes it could be nasty. 

[Andrew Dayer]

Thanks Andy and Doug. In a semi-scientific attempt to work out what's happening, I've modded another 285HV, removing and not replacing the UV blocking window over the flash tube. I can thus compare the CD window + zoom fresnel on flash 1 with the naked tube on flash 2 - assuming for no good reason that their outputs are more or less the same.

 

Two notes on modifying these flashes:

 

- safety - dangerous high voltage inside - risk to life!  No part of the inside of these units should be considered safe until the capacitor is certain to have been fully discharged and even then treat as high risk as a precaution. The internals of these units vary quite a lot so don't assume too much.

- ease of modding. The ticky part of these is supposed to be reconnecting the articulared fork arm that connects the zoom to a potentiometer near the capcitor. I found it much easier to just leave the forked arm by disconnecting at the articulaing elbow. On the other hand, I found not breaking brittle solder to be the tricky bit.

 

I have no way of quantifing the differences so the best I can do right now is take a picture of the flashes in the same frame and judge which one seems strongest. Improvement suggestions on a postcard....

Basic set up:

 

Camera: Nikon 3200 'full spectrum' (but sensitivity and UV depth not known)* at 1/200; ISO100. Pop up flash at min (1/32) power as optical trigger

Lens: Nikon 105mm f4.5 UV @ f32
Filter: Baader Venus-U (also Hoya UV / IR block & no filter)

 

* I had a conversion with a technician from ACS, a UK converter and possible converter of this body. He mentioned they replace the UV and IR filters with glass that is good down to about 330nm but not much below.

Remote shutter trigger from behind UV flash units; UV safey glasses used.

Lights:

 

Left: Alonefire X901UV 10W 365nm UV Torch (not warmed up or anything; just for interest and a reference for alignment when Baader Venus-U used

Centre: Flash 1 - Vivitar 285HV at min (1/16) power. Flash tube window from 'CD plastic' (PS?) and standard zoom fresnel. SYK-3 optical trigger.

Right: Flash 2 - Vivitar 285HV at min (1/16) power. Flash tube window and standard zoom fresnel removed and used naked or with varied window(s). SYK-3 optical trigger.

Sample pictures (camera preview JPG's at reduced resolution. No other processing or white balancing)

 

DSC_451.jpg. Flash 2 (RHS) with no windows:

 

 

 

DSC_454.jpg; flash 2 with original flash tube 'UV block' window re-installed:

 

 

DSC_452.jpg; flash 2 no flash tube window; zoom fresnel reinstalled:

 

 

 

DSC_456.jpg; flash 2 original flash tube window ('UV block') and fresnel window:

 

 

DSC_464.jpg; flash 2 with flash tube covered by SCHOTT B270*; no zoom fresnel:

* actually a micro-arraying slide I happened to so might have functional silane coating; understood to be a B270 substrate . Slide is 25mm high so doesn't fully cover flash reflector.... but just tried it for interest.

 

 

Apart from the flare and reflections (filter / front element?), does anyone have a view?

Can I do a better test with the limited equipment I've got? I don't have a flash meter but I suppose that's not intended for UV either.

 

[Andy Perrin]

I don’t think it’s a good test, taking a photo of the light sources. They just saturate the sensor, as you see above, and you can’t really tell which is strongest. Instead it might be better to take photos with each source of a neutral surface (like PTFE, perhaps at an angle to the camera to avoid saturation). Make sure the photos are not saturating with the brightest source. Then the rest will not be saturating either. Finally you can use the RAW to rank them relative to each other using the histogram peak. 

[ulf]

I agree with what Andy say above.

 

Also make sure that the distance between the light source (flash and the surface) is exactly the same each test.

The intensity decreases very fast with increased distance. 40% more distance halves the intensity at the surface. Two times the distance leave just 1/4 intensity.

 

There is another aspect when comparing the output from any light source, the illumination pattern.

That also makes these comparisons difficult.

 

A more tight beam gives more intensity locally while a more wide light pattern covering more surface will be much less intense. 

If you imagine an evenly illuminated circular light spot that can have different diameter, while containing the same power, by increasing the diameter of the spot by a factor of 2 the intensity on the surface decrease to 1/4.

[Andrew Dayer]

@Andy Perrin & @ulf  thanks both very much or taking the time and trouble to guide me whilst I scratch around in the (UV illuminated) dark.

I've just run a test series based on your recommendations and yes, its much better / more sensitive. D3200 / 105 UV / Baader Venus-U. Light to reflector (200mm square virgin PTFE) distance about 500mm; reflector to camera sensor about 1200mm. Lens defocused (infinity)..

It'll take me a while to access the NEF files properly - work is a priority, apparently. Is any PC application particularly recommended for assessing histograms? I have used LR, PS, Nikon's NX etc and also Darktable but not eg Rawdigger. I maybe overthinking here...

A highly subjective glance at the JPG previews suggests that compared to the naked flash tube with its reflector:

the original UV block window reduces the signal significantly including a colour shift. Maybe -3 to 4 stops
 

the original zoom fresnel reduces - or does its job and widens the beam but same difference - somewhat. Maybe -2 stops.
 

CD plastic similar to the zoom fresnel - about -2 stops.
 

microarray slide (B263 type) - maybe -1/2 to 1 stop
 

Alonefire 10W 365nm LED torch - about -3 to 4 stops.

Of course, its a bit hard to know how much visible & IR signal I'm getting through the Baader Venus-U; not too much I think.
 

So I guess using the glass slide is best way forward for now. I don't feel happy about the tube being unprotected from accidental finger touches or if it fails dramatically.

Is it worth thinking about using fused silica (quartz) like JSG1 or that unnecessary? D263 at the slide thickness (~1mm) transmits (SCHOTT data) 90% at 350nm falling to 0% around 310nm. Is it know if these xenon flash tubes output in the range that's being cut.

Any further thoughts or guidance welcome.

 

[ulf]

I have RAWdigger, but do not use it that much. It is likely the most powerful tool for analysing and comparing RAW-histograms.

 

My shortcut instead is the FastRawViewer from the same company that created RAWdigger.

In the preference settings you can set the exposure manipulation step length.

The finest step is 1/12EV, but I have the setting at 1/6EV

 

When I compare two files I try to match their histogram peaks by changing the exposure on one of them, while flipping between the images.

You can then read how big the adjustment has been after properly aligning at least one of the peaks.

 

If you do a WB you can also read the gain change between the RGB-channels for a proper WB. That is an indicator of the spectrum content.

Then look for the ratio between R and B. More red is a hint of deeper UV reach.

The program do not have a WB with a lasso or area, but it can use differently big groups of pixels up to 33x33.

[Adrian]

For interest: This is my "standard" kit for much of my UVR work (constantly reviewed!):

 4 x Metz 45CT flash guns. I have removed the plastic window in front of the Xenon tube (very easy!) and they are set to maximum power. I buy these from eBay for around £25 each. There is a Nikon SB24, again with the plastic window removed (more difficult!) which triggers slave cells attached to the flashes. This is also to save the electronics on the camera from the high trigger voltage of the Metz flash guns.

 

I just about manage to achieve an aperture of around f/11 at 400 ISO with a Baader U filter when they are about 12" away from the subject! This is a lot of light!

[lukaszgryglicki]

Wow :)

 

[Andy Perrin]

Wow indeed, Adrian!

--

Honestly, the histogram will be roughly the same between apps - only the bin size might be different. But for relative comparisons of the peak, that is probably not a big deal. RAWdigger is the best for this stuff, but as long as it's the RAW histogram (not JPEG or even TIFF) then it will be fine in other programs.

 

If you are not interested in UV-B or UV-C, you may want to deliberately choose a material that blocks them while passing UV-A, as a safety measure. UV-A is not very dangerous if you use safety goggles, but the shorter waves can give you skin cancer.

[Doug A]

On 9/20/2022 at 11:27 PM, Andy Perrin said:

Doug if that thing explodes it could be nasty. 

I don't see great  risk. It is only aimed at plants and rocks. I'm never in front of it. It might take out a flower, but that's about it. I'll be careful.

Thanks,

Doug A

[nfoto]

Sharp shards might fly everywhere and can ricochet.  Seeing first hand the damage potential,  I for one would never use a flash without some kind of protective cover or dome.

 

My Broncolor studio flashes have uncoated quartz glass domes. The SB-140 units have the stock cover in place, however the flash is designed for UV output by Nikon.

[Andrew Dayer]

On 9/22/2022 at 12:33 PM, Adrian said:

For interest: This is my "standard" kit for much of my UVR work (constantly reviewed!):

 4 x Metz 45CT flash guns. I have removed the plastic window in front of the Xenon tube (very easy!) and they are set to maximum power. I buy these from eBay for around £25 each. There is a Nikon SB24, again with the plastic window removed (more difficult!) which triggers slave cells attached to the flashes. This is also to save the electronics on the camera from the high trigger voltage of the Metz flash guns.

 

I just about manage to achieve an aperture of around f/11 at 400 ISO with a Baader U filter when they are about 12" away from the subject! This is a lot of light!

 

 

That is impressive - but maybe 50:50 on photographing the subject and cooking it!

I did look at following your lead with the 45CT's but they seem to be scarce now and priced at £50+ now on eBay whereas I'm playing upto £15 delivered for the Vivitars and there are plenty out there. Power-wise, looks like the Vivitars at max output (telephoto) are not much less than the CT45's and both are put in the shade by the Nikon.

 

By co-incidence I brought a new SB-24 (still bagged and boxed) earlier this week but I don't think I could bear to wreck that just yet. But does it have a significant UV output - I just assumed most modern speedlites had coated tubes??

Think out loud here, the Vivitar 285HV has a zoom head, and one of fiddly bits in removing the flash tube window is the mechanical linkage between that head and a potentiometer. The manual shows a higher guide number for the telephoto position than the wide angle position; presumably the potientiometer is varying the flash output? To answer that, I made images in the last test with the zoom in 'wide' and 'tele' (both naked tube; no windows; same postion and settings). There is no significant difference (on a sample size of one...). So what does that potentiometer do??

Thinking out loud ii. Looking at your picture of five flashes, all the tubes / reflectors are visible in each of them. Most of the output from these flashes then is not reaching the subject but lost. So I'm wondering how to make a modifier that will redirect some of the lost UV - I suppose metal foil for the reflective surfaces would be cost effective?

[ulf]

3 minutes ago, Andrew Dayer said:

The manual shows a higher guide number for the telephoto position than the wide angle position; presumably the potientiometer is varying the flash output? To answer that, I made images in the last test with the zoom in 'wide' and 'tele' (both naked tube; no windows; same postion and settings). There is no significant difference (on a sample size of one...). So what does that potentiometer do??

Thinking out loud ii. Looking at your picture of five flashes, all the tubes / reflectors are visible in each of them. Most of the output from these flashes then is not reaching the subject but lost. So I'm wondering how to make a modifier that will redirect some of the lost UV - I suppose metal foil for the reflective surfaces would be cost effective?

The higher "power" output at the tele setting is due to that the flash concentrates the light more like beam by the position of the font fresnell-like plastic.

The power content in the flash is due to the energy stored in the internal flash capacitor. The guide number is just a way to indicate the light reach for an unmodified flash.

The real energy is rarely indicated in flash specifications, What you should look for is a Ws-number. Typically I guess a on camera speedlite-type of flash is around 50Ws

The needed space and weight of the capacitor has not decreased very much over the years. A Canon A199 is around 50Ws compared to my 200Ws Godox flashes.

[Andrew Dayer]

On 9/22/2022 at 4:42 PM, Andy Perrin said:

Wow indeed, Adrian!

--

Honestly, the histogram will be roughly the same between apps - only the bin size might be different. But for relative comparisons of the peak, that is probably not a big deal. RAWdigger is the best for this stuff, but as long as it's the RAW histogram (not JPEG or even TIFF) then it will be fine in other programs.

 

If you are not interested in UV-B or UV-C, you may want to deliberately choose a material that blocks them while passing UV-A, as a safety measure. UV-A is not very dangerous if you use safety goggles, but the shorter waves can give you skin cancer.

 

Thanks Andy.

Right now I'm only interested in wavelengths 'visible' to animals and am not a big fan of self-inflicted harm so NUV is about the limit of intention. Is there such a blocking material recommended for this purpose? @nfoto  noted in another thread that xenon tubes have low output below NUV. I'm not able to quantify that for the Vivitars myself.

I'm also not a self-portrait fan so tend to trigger the shutter remotely from behind the flashes; I suppose my exposure is brief, indirect and limited by largely UV absorbing room surfaces (one should not assume...). 
 

 

[Andrew Dayer]

Thanks @ulf for both posts.

I'll have a go with FastRawViewer. Right now I'm just using LR to find the whitepoints for each image and compare the +/- compensation used. Seems quite sensitive but probably more than the error margins in my whole process.

Flash power - makes sense. Haven't seen that stated for the Vivitar but physics suggests these on camera flash units are all similar under the skin. Clearly, the zoomhead does nothing much with the fresnel window removed. I may just loose the entire assembly, fixing the potientometer position (whatever its for), and end up with something less bulky.

 

[nfoto]

Whatever flash unit you end up with, don't forget to use UV eye protection when firing the flash(es). Alternatively, run an automated set up (focus stacking) and leave the room immediately.

 

'Cooking' the subject was mentioned and that aspect is not to be neglected.  Mounting the subject onto  a heat sink, or into a water contained, can help,  I found by trial and error that using a very strong flash allows the flash itself to be positioned farther away, say 1m or more,  and if the units are powered by A/C mains recycling can be fast. The latter is important in order to avoid subject response and movement due to the UV. The flash head reflector is very important and I alternate between a wide reflector when using the Brons closer and a narrow 'umbrella-type' reflector otherwise. Old literature on close-up photography recommended 'bare bulb' flash, i.e. without any reflector, and now that is a true waste of flash output. Since so much of the emitted energy is lost, the flash has to be positioned close and that brings back the heating issue.

[ulf]

1 hour ago, Andrew Dayer said:

Thanks @ulf for both posts.

I'll have a go with FastRawViewer. Right now I'm just using LR to find the whitepoints for each image and compare the +/- compensation used. Seems quite sensitive but probably more than the error margins in my whole process.

Unfortunately the FastRawViewer is just a viewer.

 

If you happen to use a Mac for processing the UV images the RPP64, (Mac only) is one of the best pure Raw converters, with a very direct manual control of all conversion parameters.

The usage is a bit odd until you get the hang of it. The effort to learn and master it is well worth though.

[ulf]

On 9/22/2022 at 1:33 PM, Adrian said:

For interest: This is my "standard" kit for much of my UVR work (constantly reviewed!):

 4 x Metz 45CT flash guns. I have removed the plastic window in front of the Xenon tube (very easy!) and they are set to maximum power. I buy these from eBay for around £25 each. There is a Nikon SB24, again with the plastic window removed (more difficult!) which triggers slave cells attached to the flashes. This is also to save the electronics on the camera from the high trigger voltage of the Metz flash guns.

 

I just about manage to achieve an aperture of around f/11 at 400 ISO with a Baader U filter when they are about 12" away from the subject! This is a lot of light!

The Metz 45CT might be a good, but spacious, low cost path for enough flash power for UV.

They are from the same time frame as the Canon 199A with known deeper UV-output. It is likely that the CT45 flash tube also is uncoated.

The array above is energy-wise comparable to two of my Godox AD200

 

A more powerful alternative to the Metz 45CT might be the Metz 60CT-2

From what I have seen on the net all these Metz flashes operate with an actual flash voltage of 350V, not at the rated 360V.

That would give a 45CT around 100Ws and a 60 CT-2 around 200Ws

"

mecablitz 45 CL/CT-4: 1700 µF/360 V
mecablitz 60 CT-2: 3300 µF/360 V
mecablitz 402: 2000 µF/360 V
mecablitz 202: 2 x 1000 µF/360 V

"

 

[colinbm]

I have some Metz, are the capacitors still available ?

 

[Andrew Dayer]

3 hours ago, nfoto said:

Whatever flash unit you end up with, don't forget to use UV eye protection when firing the flash(es). Alternatively, run an automated set up (focus stacking) and leave the room immediately.

 

'Cooking' the subject was mentioned and that aspect is not to be neglected.  Mounting the subject onto  a heat sink, or into a water contained, can help,  I found by trial and error that using a very strong flash allows the flash itself to be positioned farther away, say 1m or more,  and if the units are powered by A/C mains recycling can be fast. The latter is important in order to avoid subject response and movement due to the UV. The flash head reflector is very important and I alternate between a wide reflector when using the Brons closer and a narrow 'umbrella-type' reflector otherwise. Old literature on close-up photography recommended 'bare bulb' flash, i.e. without any reflector, and now that is a true waste of flash output. Since so much of the emitted energy is lost, the flash has to be positioned close and that brings back the heating issue.

 

Back when I was younger and more foolish (or more fun according to my kids...) we used to demo a 250W cold light source (halogen lamp in a box and fibre optic) setting fire  to inanimate objects just from absorbtion of mostly visible light. Childish but...

Not to digress, but I should check out if I have a UV transmiting fibre set anywhere.

[ulf]

I have no idea about the dimensions of the original capacitors.

Hera are some possible alternatives I quickly found:

https://www.xenonflashtubes.com/photo-flash-capacitors/360v-1500uf-photoflash-capacitor-pulsed_102.html

https://www.mouser.se/c/passive-components/capacitors/aluminum-electrolytic-capacitors/?capacitance=1500 uF&voltage rating dc=360 VDC~~385 VDC&rp=passive-components%2Fcapacitors%2Faluminum-electrolytic-capacitors|~Voltage Rating DC

 

Another thing that breaks down over time is the battery packs. These old flashes are based on old battery technology. 

[Andrew Dayer]

18 minutes ago, ulf said:

 

... They are from the same time frame as the Canon 199A with known deeper UV-output. It is likely that the CT45 flash tube also is uncoated...

 

 

 

Just to be clear, the Canon goes deeper into UV than the Metz CT45?