What's new
Medium Format Forum

Register a free account now!

If you are registered, you get access to the members only section, can participate in the buy & sell second hand forum and last but not least you can reserve your preferred username before someone else takes it.

WAHOO 203FE back from overseas With CFV ability now

Hi Marc,

> What's the point? To prove you are right? So what if you're right? The > back is what it is. So don't buy one if it offends you so much.

This, for me at least, is an exploratory exercise...to understand what was done, why (if at all possible) and to understand the limitations. I really don’t care if I’m right or wrong, just that I understand, in the end, what is going on. Simple as that.

It’s not a matter of “offending” or not, but again, understanding.

It’s this kludgyness that has held back MF digital (backs) for many years. Luckily, it’s only taken what...more than a decade...and now it seems that some manufacturers are coming out with better integrated solutions.

Regards,

Austin
 
I understand.

And you are correct, the newer solutions are better integrated ... with Hasselblad leading the pack IMHO.
 
Austin,

"but in fact the test system could contain logic and voltage tests"

Indeed.
And since the system would 'know' the diagram, no diagram was published?

That was the point. I'm not contesting any assertion that the Databus is digital. It is.

But i do not share your confidence that a simple reprogramming could have conserved the ISO-info and add a trigger.
All that info may be put over the Databus, yes. But only if the appropriate parts are 'hooked up' to the thing to begin with.
And that is by no means certain. On the contrary: there was no need to link the bus to the release mechanism, so it's highly likely that it is not.

And, of course, you can't create a physical connection that isn't there by reprogramming a chip.
But you could rewire, and after that reprogram the chip and both retain ISO-info and add the trigger.

It looks like the solution chosen was to rewire, but only that which the contacts hook up to, i.e. disconnecting them from the Databus and bus-controller, connecting them to some switch in the release mechanism.

Good? Bad?
Simple, cheapest, effective, and on balance not causing a major disadvantage.
So i think it is a good thing.
 
Hi Q.G.,

> But i do not share your confidence that a simple reprogramming could > have conserved the ISO-info and add a trigger.

It may be a matter of how much experience you have versus how much I have with respect to digital electronics and micrprocessors and the likes and the mechanisms involved. We can’t all be experts in everything, but this is one area that I am extremely well versed in. I’ve designed systems that do exactly what we’re discussing, so I know how it can be done and what’s required to do it.

> All that info may be put over the Databus, yes. But only if the > appropriate parts are 'hooked up' to the thing to begin with. > And that is by no means certain. On the contrary: there was no need to > link the bus to the release mechanism, so it's highly likely that it > is not.

The appropriate parts *ARE* hooked up from what I can tell. All that is needed is for the CPU that outputs over the databus, which we know it does, to know when the shutter is released, and we know it does.

> And, of course, you can't create a physical connection that isn't > there by reprogramming a chip. > But you could rewire, and after that reprogram the chip and both > retain ISO-info and add the trigger.

The bus is not *directly* linked to the release mechanism, it does not need to be, nor should it be. The shutter release is simply a switch state (electronically), and the “databus” is a digital protocol. No rewiring needed, and no new connections needed...if the release switch state is known by the CPU (which it is) and if the CPU can communicate over the “databus”, which we know it can.

Why they didn’t do seems to me to be because of the mechanical cameras, not because this would not have worked. They did not want to add the electronics to talk over the databus to those cameras, which would have actually been very easy to do. They either didn’t have the expertise (again, designed by mechanical minded people) or they didn’t have the foresight to be able to come up with this as a solution.

The only way to prove one way or the other is to have the schematics of the body. Or, a circuit board that I could reverse engineer. Interestingly, Contax seems very free with their electronics, since the service manual I have has full schematics and even decent explanations of all the circuits...but Hasselblad doesn’t seem to be as willing to share this information. If you could get a schematic, or even a block diagram, that would help.

Regards,

Austin
 
Austin,

This is not a question about personal competence.
It is about knowing what functional bits are hooked up to the bus + controller.

How can you be so sure all the bits that would need to be for the conversion to be a reprogramming job indeed are hooked up?
I can't find a schematic, diagram or drawing that would suggest it is so, or not so.
And, as you know, inquiring minds do want to know.

As said before (i think) the release mechanism of 2000- and 200-series cameras consists of a bunch of separate mechanical and electrical bits. With many switches, that synchronize the mechanical and electrical funtions.
The Databus was only used to provide the metering electronics with input. And the metering electronics only 'hook up' to the time setting mechanism.
There was no need for the meter nor time setting mechanism to know anything about the release state of the camera.

So is the release mechanism linked to the bus + controller? We don't know. As you say: we need something like a diagram.

The assumed fact that the 'CFV-trigger' conversion links the databus contacts (not the bus obviously) to a mechanicaly actuated switch inside the camera would suggest there could indeed not be such a link.
Why else indeed not simply reprogram the controller?
wink.gif


So it's a we-simply-do-not-know situation we are stuck in here.
But what we do know is that - even though i can understand how, and respect that it would offend an EE's aesthetics - it doesn't matter much. Marc is right: it is not an issue to heap scorn on Hasselblad over.
 
Well well well,

I would like to know if it is possible that one who has a no-convertted 203F with "F" setted lens to try to use the CFV as the body is a 500cm ou a SWC.
 
Isidor,

You suspect the conversion isn't necessary at all, and it is all a ploy trying to hoodwink people out of their hard earned money?

You could also try the reverse, use an E-back on a converted 203. If it still transmits ISO values, we know for sure that it is a con!
wink.gif
 
Yes and no.

I have a 2000FC to repair, a 2000FCM and a 202FA working perfectly and the 110mm with I use a lot.
The prize for upgrade of the 202FA is close to the prize I got the 202FA
On literature it's writen that C lenses can not be use with the 202FA. That was not completely true. I can't trust literature which write "But, who needs film... "
It's hard to imagine that the CFV could recognize that the body is a 500CM or a 2000FCM on "C" speed if you do not set it.
At the end I cann imagine that's not hard to make a special "bulb-flash syncronisation" in a 2000FC to fire the CFV back.
At the end: In less than 2 years, a new back could arrive in witch the modification is no more needed. By the way one day I will buy a 203FE and sell the 202FA; buyers may consider that the upgrade is not an improvement because he cann not use his E-12 functionalities.
Perhaps when I will get the CFV or next, I still need film to keep wide angle ability
 
Hi Q.G.,

> This is not a question about personal competence.

I never thought it was, but some are more qualified than others to evaluate and understand based on what one's background and experience are. It certainly doesn't mean someone who has vast experience and background can't be wrong, but the odds of them being wrong in their area of expertise is certainly reduced substantially. When in NYC, I'd rather ask directions from a cabby that has been driving a cab in the city for a decade than from a guy who has been in the city for an hour.

> It is about knowing what functional bits are hooked up to the bus + > controller.

It's also about understanding the mechanisms involved. We know the CPU (what you are calling controller) is hooked up to the databus, and we know the shutter trigger is hooked up to the CPU, that's all that's needed. Beyond that, nothing more is necessary, providing we have the original source code as well as there is sufficient storage in the program memory to add this function.

If there is a program memory size problem, that is typically easily solvable with only code modifications through code optimization. If we don't have the source code, that is a different story, but not insurmountable...as I've taken over/run projects that have had this as an issue, and have always been able recreate a code base.

> How can you be so sure all the bits that would need to be for the > conversion to be a reprogramming job indeed are hooked up?

Because I have the knowledge and experience to know how to determine that with very high certainty. I have spent countless hours reverse engineering electronics for patent cases, as well as designing with the same technologies that are incorporated in this camera. I solve problems like this for a living, and have been doing so for 30 years now. This experience gives me the ability to evaluate the design by using the written material provided, as well as the camera operation it self. I am given black boxes with little to no documentation all the time and told to interface to them.

If the CPU wasn't hooked up to the databus, then it couldn't read the ISO from the back...and we know it can. If the shutter trigger wasn't hooked up to the CPU, then the CPU would not know when the shutter was triggered, and we know it does.

> The Databus was only used to provide the metering electronics with > input. And the metering electronics only 'hook up' to the time setting > mechanism.

That's not correct. It seems perhaps you are misunderstanding how the camera's electronics function, on a base level.

A quote from the 205TCC brochure:

"The camera's brain is a microprocessor or CPU, which calculates all the necessary data and controlls all the functions."

You can view the ISO setting from the back in the camera's display. That display is controlled by the CPU. It's clearly stated in the 205TCC brochure the CPU is for (at least) exposure determination, and in order to do that exposure determination, it needs the inputs from the lense and back. It also obviously has the switch states (side dial/button, and shutter button) as inputs, as well as the display as an output.

We know the shutter is linked to the CPU as well, since the CPU is what determines the shutter speed, as well as it displays the speed in the display. You can also program the self timer delay, and over/under exposure limits as well as a meter offset via the CPU.

The CPU *IS* the "metering electronics" *AND* the "time setting" mechanism, as well as the controller of other camera functions. Again, this is clear from not only their literature, but from camera operation.

> So is the release mechanism linked to the bus + controller? We don't > know. As you say: we need something like a diagram.

We *DO* know. The release mechanism is absolutely linked to the controller, since when you press the shutter release, the display changes, and the display is controlled by the CPU, and so is the shutter speed. The databus is controlled by the CPU as well. There is no need for the release mechanism to be linked directly to the "bus", which I assume you mean databus. It is "linked" through the CPU.

> The assumed fact that the 'CFV-trigger' conversion links the databus > contacts (not the bus obviously) to a mechanicaly actuated switch > inside the camera would suggest there could indeed not be such a link.

It absolutely does not. If you read the brochure and had experience with the operation of the camera, this would be obvious. Again, as I've outlined, it's %100 certain that there is this link, or the camera could not function as it does.

> Why else indeed not simply reprogram the controller?

As I've stated before, my speculation is because of the mechanical cameras. They do not have the electronics in them to drive the databus. That does not mean they couldn't have been retrofitted with some small circuitry to do this, but that is not how they chose to do it.

Other reasons are they had no one qualified to do this, or they didn't have the source code and/or the development systems setup any more. At this point in time, only they know this answer for sure.

> But what we do know is that - even though i can understand how, and > respect that it would offend an EE's aesthetics - it doesn't matter > much.

It's not an aesthetic issue. For me, it is a function issue. Not having the ISO (and the development compensation) from the back looses a critical function that I use the 205 for, especially since I use the Zone mode with the development compensation very frequently, and it is a major feature of the 205, and the reason I bought it.

> Marc is right: it is not an issue to heap scorn on Hasselblad > over.

No one is heaping scorn at Hasselblad at all, simply trying to understand what limitations they are imposing, and why. That understanding may not mean anything to you (or Marc), but it does to me. Others have their "nits" of "understanding" and I have mine. I don't begrudge others for theirs, and others should not begrudge me for mine.

You just won't give me any credibility, will you. I don't take it personally, mind you. This is not in any way meant as a slight, but from your responses, it appears to me you don't have much experience with designing digital electronics. Please correct me if I'm wrong. If true, I'm not sure why you wouldn't give credence to someone who does have this experience.

Regards,

Austin
 
Austin,

"and we know the shutter trigger is hooked up to the CPU"

No. We do not (!) know that.

We do know that there is a switch in the release to activate the electronics.
But that does not mean it is linked to the release mechanism, which still, and despite CPU, is mechanical in 200-series cameras.

The time setting mechanism in the 2000-series is electrical. And so it was too in the 2000-series, so again that does not mean there must be a CPU involved, except as a producer of regulating input for the mechanism (something the resistor array did in the 2000).
And it's not something that would need to know when therelease is pressed.

Unless we see a diagram, we just do not know.
 
Hi Q.G.,

> "and we know the shutter trigger is hooked up to the CPU" > > No. We do not (!) know that.

"WE" may not, but I absolutely do! Again, if you had one of these cameras, you would be able to experiment for your self and see.

There are three positions for the shutter release. Not pressed, half way pressed, and fully pressed. If I put the mode selector in "Pr" it displays "Pr 10s". When depressing the shutter button half way, the display changes to displaying the aperture setting and the shutter speed. While in this half way position, when moving the camera around to varying lighting conditions, the shutter speed changes in the display. When the shutter button is fully pressed, it locks the shutter speed so it does not change, and it does not change when moving the camera around to varying lighting conditions.

Clearly three distinct display condition, each based on one of the three positions of the shutter release. The display is unquestionably controlled by the CPU, so the shutter release is unquestionably "hooked up" to the CPU.

Regards,

Austin
 
Hi Timothy,

> My god, guys! Give it a rest. My inbox begs you.

Ha ha...once Q.G. says "uncle" ;-)

But, seriously, I think we're almost at an end. I've substantiated the interface works as I said it does from stem to stern...and the only thing left is getting the schematics. Which, of course, would be wonderful if we could.

Regards,

Austin
 
Timothy,

Now we have a discussion on a discussion forum, and it's not allowed?

Austin,

What does the display do when you prerelease the camera?
 
Hi Q.G.,

> What does the display do when you prerelease the camera?

When you press the pre-release, the display goes to the state where the shutter release is depressed half-way. When you unpress the pre-release, the display goes back to the state where the shutter release is not pressed.

I believe it's also reasonable to conclude that the self timer is controlled by the CPU.

As a note, the 205 service manual gives some more clues. On p. 3:5, they talk about being able to release the magnets on the main circuit board:

"The magnets can be released from the main circuit board when troubleshooting."

I think it's also reasonable to conclude that these are the shutter release magnets. One for first curtain, and one for second curtain. If you look at the picture of the "magnet" assembly shown on p. 3:5, and go to p. 17, you'll see two of those devices shown, one above each of the shutter gears. They are not listed as spare parts in the parts list (no designation number for this assembly is given), as it states on p. 3:5.

So, we know these magnets can be released from the circuit board, and if they are in fact the shutter release magnets (which they appear to be), they are controlled electronically on the main circuit board. They are also going to be electromagnets, not simply magnets.

One other interesting note in the service manual on p. 1:1:

"Electronically controlled mechanical focal plane shutter with (203FE/205FCC) release solenoid system."

As we know, solenoids are electromagnetic devices...so perhaps this was just a "confusing" translation.

See p. 19 as well. I think the circuit board they show is the one that interfaces to the shutter release button, and some other mechanical I/O. One they call the "sync switch assembly" (#8), and another the "late release switch" (#4), but it's the one that has no designation and has four numbered wires going to it that I'm interested in. It almost looks like it could be a pair of electromagnets...that could coinside with the assembly referenced above.

Regards,

Austin
 
Hello Austin,

I agree that the display may hold the clue.

The 2000-series cameras have the same mix of electric and mechanical bits involved in the release mechanism and procedure. The magnets mentioned are what 'the mechanical guys' contributed, helping the thing along again.
They (the magnets) in the 2000-series can be released from the circuit board too, and these cameras do not have a CPU.
So i don't think this in particular tells us much.

(Hasselblad once 'invented' a remote release cord for the 2000 FC. It plugged into the battery box, and released the camera 'from the circuit board'. It never made it to the shops, since there was no way to reset the camera and wind film remotely too.
wink.gif
)

The 'late' and 'early' switches, by the way, have to do with 'C' and 'F' modes.
 
Hi Q.G.,

> The 2000-series cameras have the same mix of electric and mechanical > bits involved in the release mechanism and procedure. >... > They (the magnets) in the 2000-series can be released from the circuit > board too, and these cameras do not have a CPU. > So i don't think this in particular tells us much.

I think it tells us a lot. The timing of the shutter is done by some electronics in each camera, we know and agree on that. In the 2000 series it is by analog control (we know and agree on that), and in the 20x, it is done under digital control. We know that is so in the 20x, since it is verly clearly stated in the brochure (and through use) that the metering can control the shutter, and the CPU is what determines the metering. And, the CPU is digital... So though we apparently don't agree on this entirely, I think, again, the documentation and operation bares this out.

How, exactly, do you believe it works from the shutter button to the shutter release?

> The 'late' and 'early' switches, by the way, have to do with 'C' and > 'F' modes.

I hadn't associated those, thanks.

Regards,

Austin
 
Hi Q.G.,

> I agree that the display may hold the clue.

The display tells us that the shutter release button state is known by the CPU, which is one of the things you were questioning. I believe this is not questionable now, given the evidence.

In my book, the only remaining question is, and it's kind of two-fold...does the CPU actually trip the shutter, or is it done electro-mechanically. And, if not, what is the timing relationship between the CPU knowing the shutter is tripped and the shutter actually tripping.

The easiest and most accurate way to operate a shutter if you have a CPU in the device already...is to take the shutter button as an input to the CPU, and since the CPU knows the shutter speed, it would output a release of the first magnet, therefore tripping the first shutter curtain...and then, after the appropriate shutter time, release the second magnet, therefore tripping the second shutter curtain.

The CPU has to have a clock, that is part and parcel of how it operates, and 1/2000 is very slow in CPU time. It is only 500 microseconds. Even a really slow CPU, say at 1MHz, or 1 microsecond per cycle, would easily be able to handle timing 500 microseconds, either by using a built-in timer (which most have) or by simply counting. This is very easy to do. The CPU is doing nothing else during this time anyway, so even if it was a dedicated timing loop, no other operation would be effected.

Regards,

Austin
 
"The display tells us that the shutter release button state is known by the CPU, which is one of the things you were questioning."

I wasn't.
I was (still am) wondering whether the controller knows the shutter is open.
And that's not splitting hairs, it's not the same. You can - as every 500-series knows - let go of it before the shutter has closed. Or keep the release button depressed long after the shutter has closed again.
That's why the behaviour of the display relative to the position of the release button might hold a clue, but is not proof.

The answer to one of the two parts of your remaining question: the shutter is tripped by mechanical bits. That's why it still works in C mode even when there is no power.
The timing of the shutter is done by the electrical bits.
But whether that means it is done by the controller, or (like in the 2000) by some other bit?

The speeds are adjusted using a potentiometer. Could indicate that some means are necessary to adjust the other electro(-mechanical) bits to keep up with the controller's precision. Or that the timing is not done by a CPU.
 
Hi Q.G.,

> "The display tells us that the shutter release button state is known > by the CPU, which is one of the things you were questioning." > > I wasn't. > I was (still am) wondering whether the controller knows the shutter is > open. > And that's not splitting hairs, it's not the same. You can - as every > 500-series knows - let go of it before the shutter has closed.

It only needs to know when it's released, so it can send that to the back.

But, it does in fact know when it's open, if it knows when it's released. The CPU knows the shutter speed, and knowing when the shutter first opens, it can simply count from there. So, that is not an issue.

> The answer to one of the two parts of your remaining question: the > shutter is tripped by mechanical bits.

The shutter curtains do obviously open and close, like the barn doors on the 500 do, when in C mode, but there is no shutter timing circuitry per se. But, that does not mean that when not in C mode, they are not tripped fully electronically. That conclusion can't be drawn from any of the information we have to date as far as I can tell.

> The speeds are adjusted using a potentiometer.

From what I can tell, only one speed...1/2000...is adjusted by a pot. At least that is what the service manual says. But, that is no indication as to how the timing is done. It can simply be adjusting a time base. But, this seems strange to me...since if you have an oscillator for a microprocessor, it is handily well within any tolerance needed to accurately produce 1/2000.

But, to go back to the initial supposition...all the CPU needs to know is when the shutter release is fully pressed, it doesn't matter how long it's pressed for. And we know that the CPU does know when the shutter release is fully pressed (to trigger the shutter). Whether the timing is "correct" is another issue.

Regards,

Austin
 
You must log in or register to reply here.
Back
Top