Breathing life into an old flashgun... Part 1

[Vortex]

Having recently returned to photography after a break of almost twenty years I'm in the process of 'reassessing' the gear I have.

I have a circa 1991 Metz 45 CL-4, the pre 'digital' version, which with an SCA 346/2 AF served me well. Alas, today it's unable to 'talk' to my Nikon D300 and can only be used in Auto or Manual mode - so, it seemed ripe for a few modifications. However, first things first ... charge the battery pack.

Disclaimer: This is not a DIY step by step guide. If you intend to modify any equipment of your own, ensure you fully understand what you are doing and are aware of any risks involved.

Part 1.

First off, power to the beast. Originally it used either a NiCd battery pack (45-40), replaceable alkaline cells in holder 45-39, external high capacity NiCd pack 45-45, or mains adaptor N 22. All of which have been discontinued.





This is the original NiCd pack, in remarkably good physical condition considering it's age, with no leakage from any of the cells.


I was not really expecting it to hold a charge after having been left for so long. The Metz 703 charger is of the 'fast' type and I didn't want to use that for initial charging, a much slower charge rate will be more effective in reviving the NiCd cells. (There is much written on the web, both fact and fiction, regarding NiCd cells... I've spent the last forty years or so working in the electronics industry, I sometimes laugh, sometimes almost cry.)

Unfortunately the 703 charger supplies AC; there being four diodes forming a bridge rectifier within the battery pack itself, together with a small filament lamp which serves to both limit charging current, and to give an indication of charging. I don't require the diodes or lamp and these were removed from the battery pack.

This is not too technical, the maths is minimal, don't be put off.

The Metz 45-40 NiCd pack contains six 1.2V 500mA/h cells to give 7.2V, to trickle charge these a voltage of 1.41V per cell at a charge current of 0.05C is required. Which gives us 8.46V at 25mA. Sorting through various mains adaptors lying around I found a 10V 1A adaptor, with the added bonus of it being a switch mode supply, ideal!

Let's calculate the dropper resistor needed... we need to drop 1.54V and limit the (charging) current to 25mA. Pay attention at the back, who remembers Ohm's Law? V=IR, therefore R=V/I, we need 1.54/0.025 = 61.6 ohm resistor. Opt to use 68R as a 'preferred' value, it needs to dissipate V²/R watts of power which works out as 0.035W, let's play safe and use a 3W wire wound resistor.




The modified mains adaptor with dropper resistor in series, simply insulated with heat-shrink sleeving.




Initial charge ... hmm.. I wonder

The coulometric charging efficiency of nickel cadmium cells is about 50% at these low charge rates, so to fully charge the cells will need approximately 500/25x1.5 hours (30hrs). At this charge rate it is safe to leave cells on permanent charge, I initially gave them approx 36 hours.

Into the flashgun and ... 14 (yes fourteen) full power flashes ... wow! These cells have not seen a charge in many a long year, I was actually quite surprised! After a second 36 hour charge the pack delivered 17 full power flashes, after a few more charge cycles I was obtaining 34 full power flashes. Metz quote 50 full power flashes from those packs - well, they are only 500mA/h cells. Not bad, 68% of original capacity!

At some time in the next few months I'll open up the battery pack and replace the old cells with nickel metal hydrides.


To follow...

Part 2 - Let's go mains supply instead.
Part 3 - Ditch those proprietary connectors / High capacity battery pack.
Part 4 - Fully variable power (flash duration) control.

[Colin Southern]

Hi Paul,

I know a lot of people find these kinds of projects fun (including me), but at the end of the day I can't help but think that people are - nostalgia aside - far better off just chucking it back in the cupboard and buying something that's compatible; it'll make their photography far far easier.

[Kodiak]

…nostalgia aside - far better off just chucking it back in the cupboard

I have a few of the 45CT5 that, at the F4 time, coupled with their TTL cable.
These high-tech guns (dinosaurs today) are working very well and still use them.

The only thing is the electrochemical compound in use then: NiCad! The best
and, at the same time, the most dangerous cell to use with flash… Paul is right
to warn of potential danger! I loved those NiCad's, and the memory effect
could be taken care of when recharging the right way.

[Vortex]

... but at the end of the day I can't help but think that people are - nostalgia aside - far better off just chucking it back in the cupboard and buying something that's compatible; it'll make their photography far far easier.

Hi Colin

Undoubtedly, if one's interest is purely in photography then I would not disagree.

Personally, I tend to find myself 'merging' my interests. In this particular case is was not so much compatibility with modern equipment I was after, more putting a perfectly good, relatively high power, flashgun to some use. I'm also a 'mean skin-flint rat-bag' - "chucking it back in the cupboard", no way .

[Colin Southern]

and the memory effect
could be taken care of when recharging the right way.

Just don't make contact with the storage capacitor or one may experience a different kind of "memory effect"

[Colin Southern]

Hi Colin

Undoubtedly, if one's interest is purely in photography then I would not disagree.

Personally, I tend to find myself 'merging' my interests. In this particular case is was not so much compatibility with modern equipment I was after, more putting a perfectly good, relatively high power, flashgun to some use. I'm also a 'mean skin-flint rat-bag' - "chucking it back in the cupboard", no way .

To be honest, I've done the same many times. Usually works out hopelessly uneconomic in my case (ie I waste a week on it and still have to buy a new one at the end of it).

[ajohnw]

I take it you will check the trigger voltage before trying it on the camera? Nikon cameras are still pretty good in that respect - I think. Isolators are available on ebay. I use one just to be sure. They usually have a normal flash shoe connection and an X-sync type socker. Most cameras these days only accept rather low trigger voltages especially Canon. Old variable power flashes with switches on the back rather than holes that rotate over the sensor are often ok but ......... They are sometime way too powerful for modern ISO's too.

John
-

[Vortex]

The only thing is the electrochemical compound in use then: NiCad!

Hi Kodiak,

I'm not trying to defend them, but, actually NiCd batteries are nowhere near as 'bad' as is generally thought. Cadmium is certainly a toxic substance, however the amount used in a typical AA cell is very small; it's in the form of a cadmium plated mesh used for the anode of the cell. A large quantity of electrical/electronic components use toxic materials in their construction; arsenic, mercury, and beryllium oxide spring to mind. Used and disposed of correctly they are perfectly safe.

I loved those NiCad's, and the memory effect could be taken care of when recharging the right way.

Charged at a moderate rate, discharged to around 1.05V per cell, NiCds have a very long service life.

A lot of the problem lies with 'designed to a price' equipment - basically allowing the cells to discharge below 1V. 'Fast' chargers don't help the situation either, especially the poorer designed ones that don't adequately detect when the cells are fully charged.

[Kodiak]

just don't make contact with the storage capacitor or one may experience a different kind of "memory effect"

right!

[Vortex]

I take it you will check the trigger voltage before trying it on the camera

Hi John,

From the horses mouth, the Metz 45 CL-4 has a trigger voltage of <15V. Yes, I have confirmed that.

Be very wary of claims for trigger voltages that are circulating on the net for various flashguns. Without knowing how these were measured the accuracy is *very* doubtful. A trigger pulse will typically be of much shorter duration than the sampling time of a digital mult-meter, perhaps the best method is to observe the pulse using a 'scope.

[Kodiak]

Hi Kodiak,

I'm not trying to defend them…

…but you should… as no other compound is more suitable for flash use!

And, once again you're right: —Used and disposed of correctly they are perfectly safe.

Good luck! …already waiting for part 2…

[ajohnw]

Hi John,

From the horses mouth, the Metz 45 CL-4 has a trigger voltage of <15V. Yes, I have confirmed that.

Be very wary of claims for trigger voltages that are circulating on the net for various flashguns. Without knowing how these were measured the accuracy is *very* doubtful. A trigger pulse will typically be of much shorter duration than the sampling time of a digital mult-meter, perhaps the best method is to observe the pulse using a 'scope.

I think they sit high and the camera drags them down so any DVM should measure them. I can''t remember what Nikon DLSR's can take but it is in the manuals some where. The only worry if within that is current which made think ok but fit one of the adapters that also isolate just in case. I understand these use an opto isolator that has a large voltage margin.

Like your signature. I'm on OpenSuse 12.3 KDE. . Don't get into arguments with Colin about Adobe products. It seems we could run them but ............ My photo software list is too long to include. I use which ever is appropriate. Fotoxx usually figures some where. Rawtherapee too and of course the GIMP when I need layers. Actually as it may be of interest I can list them



Corel is no more I trialed it.

John
-

[Vortex]

I think they sit high and the camera drags them down so any DVM should measure them.

Hi John,

Perhaps I should have explained a little more... long version:

The standard 'PC' flash sync (1/8 inch co-axial connector) uses the convention inner is positive with respect to the 0V/Ground outer. Taking positive to ground, as you said, fires the flash.

So far so good, whilst the flash is powered and in a 'ready' state there is a steady +ve voltage on the inner pin, yes this is easily measured; albeit one may trigger the flash in the process .

In 'the good old days' it was physically a switch within the shutter mechanism that made the connection; voltage was (relatively speaking) irrelevant, as indeed was polarity.

It is what actually happens at the moment of flash trigger that is of real interest. I first discovered this back in the 90s when I did some work on studio flash units. I'll try and dig out some of the photos of the oscilloscope traces I had, if I still have them. In the meantime a rough sketch will have to do.




Initially the trigger voltage rises as the flash approaches ready state, at the point of firing - dependant upon flash unit - there may be a very short, in the order of microseconds, pulse. This in all the cases I've seen includes a negative excursion, the trigger voltage then drops to zero before rising again. It is this uS pulse that can't be measured by a DVM, but can clearly be seen on a 'scope. I didn't investigate the circuitry of the flash units that exhibited this behaviour, so won't speculate on the cause.

Whilst the switching (in camera) was mechanical these spikes did no harm. Modern day cameras invariable use either an OC transistor or a thyristor/triac to switch; the latter being far more robust. To put this more into context, as my sketch is very much *not* to scale, I've seen a steady state trigger voltage of around 30V with a +/- spike of almost 300V !

This is very much down to the design of the trigger circuitry within the flashgun, the Metz I have has no visible spike when it is triggered. On the other hand, I've also an old Vivitar 283 which spikes to around 200V.

I can''t remember what Nikon DLSR's can take but it is in the manuals some where.

Yeah, think it's in the techy bit at the end... I know the Metz is OK to use with the D300, but I use an opto-isolator with the Vivitar.

I understand these use an opto isolator that has a large voltage margin.

Most have an IO isolation of at least a couple of kV. If you were really unlucky and had an old studio flash there's a possibility you could zap the triac in the isolator, but the input side, ie the camera, would be OK...

Personally, I'd use an isolator with any non oem flashgun unless I satisfied myself of how the trigger voltage behaved.

I'm on OpenSuse 12.3 KDE

a fellow 'nixer ... nice to meet you! Not on 13.1 then? Mind you 13.2 is only just around the corner, November release target I think.

Don't get into arguments with Colin about Adobe products.

Adobe? Who're they then ...

[Peeshan]

Hello Paul
Very interesting stuff, I wondered wether it wouldn't be easier to modify it so it would accept modern AA batteries instead of the original battery pack? Like turning the battery pack into a cell holder Or maybe it wouldn't make any difference?
I have zero knowledge about this stuff so this is a real question, not a suggestion/critiscism

[ajohnw]

Interesting. I haven't been near thyristors for a very long time Last time was a 30+Kw chopper for an electric vehicle and i can't remember for the life of me if they need a diode across them for spike clamping. I'd have to draw one using 2 transistors and think. I do have a couple of scopes about if I get interested again.

The tempting aspect of flashes for me is TTL and converting one make to another. Especially older TTL to modern retaining
TTL.

My most recent flash conversion was simple - nice little camera powered flash for macro work



The aerial will be an alternative mount to make up a flash bracket at some point.. I bought some black velcro last week. White looks a bit naff.

Step 2 will be down to it covering too wide an angle so I'm hoping to narrow the beam so that I get some sensible power levels for telephoto use at up to say 4m. Sort of longer distance macro.

I thought I would run 12.3 till 13.3 came out but releases seem to have slowed down.

John
-

[Vortex]

I wondered wether it wouldn't be easier to modify it so it would accept modern AA batteries instead of the original battery pack? Like turning the battery pack into a cell holder Or maybe it wouldn't make any difference?

Hi Pierre,

This is something that I intended to do at a later date.

The existing NiCd pack can be easily dismantled and NiMH cells used instead, however there are a few other changes that would be required. Apologies for the very poor quality image, this the 'circuit diagram' for the existing NiCd pack.

The filament lamp (Ro 2001) and the bridge rectifier (D2001-D2004) would need to be removed, (which I've done already as part of the initial modification.) A suitable charger would then be required, there is quite a wide choice there, ranging from building one from scratch, modifying an existing mains adaptor, or more probably simply buying a charger. 7.2V is a very common pack voltage, all would be needed then is to arrange a suitable plug/socket to connect the charger.

For the actual cells, something along the lines of these: http://www.batteriesplus.co.uk/acata...d_Battery.html Don't consider untagged, there is no way you can easily join the cells.

Metz did produce a NiMH pack, albeit at a *very* inflated cost.

Using the 45-39 battery holder it would also be possible to use NiMH cells, but the holder would need modification as the battery voltage would be different, 7.2V as opposed to the 9V which 6 alkaline cells would give. (There are three contacts within the Metz flashgun, common, and one each for the two different voltages.)

[Vortex]

30+Kw chopper for an electric vehicle

Low power stuff then ... bet that didn't connect with a bit of 0.75mm² cable.

The tempting aspect of flashes for me is TTL and converting one make to another. Especially older TTL to modern retaining TTL.

Interesting thought that, but not straight forward by any means. Flash to camera communication used to be very simple handshaking along the lines of 0V or +ve (generally battery) voltage, reverting to logic terms, 0 or 1. Flash ready, fire, quench, etc etc... easy peasy.

Now I can only speak for Nikon kit here, but I guess it's pretty much the same all around. The simple handshaking has gone, the auxiliary contacts on the hotshoe are now used for data, and clock. There's a fully fledged data packet exchange between camera and flash. I'll bet a penny to a pound that each manufacturer has their own proprietary data exchange.

There's some interesting stuff here http://dptnt.com/2010/04/nikon-flash-interface/ to give you an idea.

My most recent flash conversion was simple - nice little camera powered flash for macro work

Nice, like it!

I thought I would run 12.3 till 13.3 came out but releases seem to have slowed down.

Yeah, 13.2 was purposely delayed to allow for a few more fixes. There's quite a lot going on at the moment in other respects as well, KDE frameworks 5 for example. If you want longer term stability maybe consider going to 13.1 evergreen.

[ajohnw]

I usually run well past the security updates - KDE framework I started with KDE 3 a long time ago. Version 5 when the current one has problems - indexing everything etc might drive me to Gnome or one of the simpler desktops.

Interesting link. All coms like that are usually a variation on Phillips I^2 C. The biggest problem is obtaining a logic analyser that will monitor and hopefully show the results in bytes. I've never been any good at reading hex as the bits stream out. One way is to roll your own especially if it's possible to find out which variant is being used by using a uP with the same interface. I've never had to interface one variant to another but they are all very similar. I've also hand cranked some but with bi directional lines a lot of care is needed. They don't leave much time to tri state port lines.

I wonder what goes on in the simpler guns. Add zoom, multi flash heads and high speed flash etc possibly rather a lot. But maybe not as much as some might think on the simple ones. The Olympus one I posted might just uses wires to the flash head and is controlled fully by the camera just as built in flashes are. I didn't lift the board out to see what was on it but it has 5 wires going to it, 3 heavier and 2 lighter so it might have full comms. I used the 8 core wire model train people use and doubled up for the 3 heavier wires - guessing that the voltages aren't high. I should have measured that really.

I informed Oly's pet uk pro photographer that I own the intellectual property.

John
-

[Vortex]

... might drive me to Gnome or one of the simpler desktops.

Haven't looked at Gnome for quite a while now. The simpler/faster/lightweight desktops are OK... but... you'll still want to use the current (photographic etc) applications, so virtually all of the core KDE/Gtk libraries will still be there... horses for courses, whatever rocks your boat. Xfce might be worth a look if you've not done so already. Still can't beat the command line though for getting things done... often find myself dropping down to a VT, bit of a command line junkie.

Interesting link. All coms like that are usually a variation on Phillips I^2 C ...

The article I linked to there looks to have died, the last post from the author was April 2010. It's certainly a serial data bus of some description, whether it's an existing standard or some proprietary design I've no idea. It's not something I'll be investigating, at least I don't think it is...

Typing as I think, not checked any of this...

I²C is supported by linux in the LM sensors code. Pretty sure the I²C tools has been split, lots of development stuff there. Maybe grab the tarball and have a look.
Sure I've seen I²C to USB adaptors.
Bit of jiggery and maybe possible to use a PC as the analyser...

Dumb flash. Simple: Ready, Fire, Quench. How about controlling it with something like the arduino which does the high level communication with the camera.

Anyway... probably have better (sorry, more important) things to do.

[ajohnw]

Anyway... probably have better (sorry, more important) things to do.

Me too and in my case after 20 years of doing that sort of thing for the automotive industry I decided I had done enough. Maybe one day I might get tempted but I very much doubt it.

John
-