NikonGear'23
Gear Talk => Lens Talk => Topic started by: Bjørn Rørslett on March 20, 2017, 22:22:48
-
The 'Dandelion' has got itself a mixed reputation over the years. Quite deserved according to today's sad experience.
I was busy shooting the evidence that blur circles are independent on format, when my 50/1.2 Nikkor AIS suddenly acted up and the camera went dead. Turned out the CPU pins of the Dandelion all of a sudden were shifted or torn off.
This lens couldn't hardly have been CPU-modified without the efforts of my friend Erik [aka 'Dr. Lens']. The contact pins were inserted through holes in the bayonet and I had hopes this would protect them from undue wear. This bliss lasted for two years.
Before and after close-ups. The Dandelion really is a fragile design.
-
Looks like an awesome job done by Doctor Lund.
Most people that do these conversions wouldn't have attempted such a modification.
A shame the quality of the Dandelion hasn't lived up to expectations.
-
I had my 80-200mm f/4 chipped by legacy2digital and one of the "pins" sheared off in a similar manner. A second chip also failed, but not for any obvious physical reason. Both times I ended up being out the cost of 1-way shipping.
One of the nasty side effects of both failures was that my IR-converted D70 still thought it had a CPU lens attached, so I couldn't even use it in unmetered manual mode.
-
Looks like we need to do another implant :D
-
I have, on my own or in co-operation with Erik, CPU-modified a great number of lenses over the years. Must be well over 100 units by now, plus over the years I have sold more CPU kits than that for users to modify their own lenses.
As original brand new Nikon spare parts are used for the contact block, the mounting guard on the contact block is solid metal, and bespoke screws keep the components firmly in place, I have yet to see a failed modification using these kits.
The snapshot below is the outcome from one of my many trips to visit Erik. The CPU modification really makes using all those Nikkor gems from the past a breeze.
I'll have to discuss with Erik what is possible to do with the 50/1.2. Oops., he answered while I posted. Putting this on a future to-do list, then.
-
Sorry about your mishap, Bjørn.
If this Ais 50/1.2 is the ONLY lens among the ones you use that cannot transfer the lens info to the body, you could simply register the 50/1.2 lens to the body. The "unchipped" 50/1.2 will always be correctly recognized as it is without switching between different registrations of more than one MF lenses.
That would be the easiest workaround.
-
One of the Nocts is still untouched. CPU-modifying it is quite risky, although Erik has done this several times. It was acceptable when he did the CPU modification on my old Noct AI (purchased for the princely sum of 150 USD by a stroke of luck). However, the newer one not only is AIS, it also cost me a bit more :(
Thus I have more than one lens of f/1.2 speed with no CPU, if the 50/1.2 remains in the current sad state. I have to pry off the Dandelion so the lens won't fry my cameras as well. However, knowing Erik, he is making plans for this contingency already. What are good friends for if not in a time of despair.
-
Oh, so long as you take BOTH Noct and 50/1.2, the workaround won't work... :(
-
Mishaps are bound to happen as long as settings are to be remembered and entered manually. The CPU project aimed to eliminate that source of error.
Having both lenses concurrently in the field occurs more frequently than you might think.
-
Wait! You say that "one of the Nocts is still untouched." Then you should be able to refrain from using that untouched Noct? Or, do you even need to switch between different Nocts? :o :o :o
-
The vertical sides of those pins are at fault. Looks like they are for a design which is intended for loads along the axis of the pin rather than the rounded contacts used by Nikon. Bad design indeed.
-
Wait! You say that "one of the Nocts is still untouched." Then you should be able to refrain from using that untouched Noct? Or, do you even need to switch between different Nocts? :o :o :o
'Untouched' as never being the victim of a Dremel :D
My Nocts are AI and AIS, respectively, and have subtle differences in their rendering. So yes, switching Nocts is a reality.
-
The vertical sides of those pins are at fault. Looks like they are for a design which is intended for loads along the axis of the pin rather than the rounded contacts used by Nikon. Bad design indeed.
You correctly identified the main reason why the Dandelion pins shear off. They are spring loaded, but exert nevertheless much more pressure onto the camera contacts than do Nikon's design which also is favoured by much more robust solid pins with rounded tips.
-
The vertical sides of those pins are at fault. Looks like they are for a design which is intended for loads along the axis of the pin rather than the rounded contacts used by Nikon. Bad design indeed.
Yes and the extremely soft and thin materials for the tubes just makes it even worse :o :o :o
-
i was in the impression that the contacts are made out of solid copper pegs :o :o :o
I have an idea that will do away with this but I am so busy I haven't even talked to my engineer friends about this project!
-
No, they are hollow inside the contact block so they can contain the screw springs, similar design as the contact pins of the 'semi-AF' TC-16A just much more vulnerable,,,,
-
An obvious remedy would be to build up more with hard epoxy around the tubes to provide better support around them, but difficult to carry out without getting glue inside the tubes...
-
The tubes have to be able to move in- and outwards, so that's not feasible I'd guess.
-
An obvious remedy would be to build up more with hard epoxy around the tubes to provide better support around them, but difficult to carry out without getting glue inside the tubes...
The issue is that with vertical walls on the pins, a sideways load does not translate into an axial force which would depress the pin.
-
It is my view that the Dandelion is built too fragily. The tiny springs and contact pins are made of ultra thin "brass" and the black housing is made of friable plastic. It works well with very careful handling but cannot stand up to what I deem normal wear and tear.
-
.... you could simply register the 50/1.2 lens to the body. The "unchipped" 50/1.2 will always be correctly recognized as it is without switching between different registrations of more than one MF lenses....
You only get limited metering abilities, the CPU allows more metering functions.
It is my view that the Dandelion is built too fragily. The tiny springs and contact pins are made of ultra thin "brass" and the black housing is made of friable plastic. It works well with very careful handling but cannot stand up to what I deem normal wear and tear.
+1
I had a Dandelion on my 500/8. Firstly the housing started to break away at the leading edge, then the pins at that same leading edge started to shear off.
Lens got stuck on the camera once the entire Dandelion body distorted enough so to remove it, it had to break the flimsy Dandelion body up.
Luckily no damage to the D800.
It was then an ordeal to remove the epoxied reside off the rear filter.
(for those that don't know, the 500/8 uses a rear filter where the Dandelion is glued too. No mods to the lens mount required at all, so it's a good intro into CPU modded manual lenses)
I was always extremely cautious/careful in mounting the 500/8 to the camera too.
FWIW: the main reason I chose the Dandelion for the 500/8, apart from the ease of install, was that the Dandelion allows the camera to set trap focus.
Focusing with the 500/8(or any long manual lens) is difficult as we all know even with a fine grained focus screen(which becomes dark at f/8! :D)
The focus trap experiment worked a treat, and apart from my unstable handholding ability, the images shot with focus trap mode enabled were as sharp as I've ever got with the 500/8 ..
summary: I like the Dandelion CPU for it's ability but will never get another due to the fragility. I also wanted to mod an old 300/2.8 Tamron(adaptall) lens I also have. But will not.
If a solution to the cheaply made Dandelion body can be worked out, I'll definitely consider them again.
ps. the other handy/advantageous property of the Dandelion CPU is the ability to program/reprogram it(via the camera). It's awkward, but doable and a lot more flexible than using the camera body selection for non CPU lenses.
The advantage with this programing system is (for example) you mount a Dandelion to an old extension tube of some kind and use various lenses on that extension tube .. you can obviously set whatever lens factors required for the various lenses.
Looking forward to what Erik can come up with.
-
The issue is that with vertical walls on the pins, a sideways load does not translate into an axial force which would depress the pin.
I think this comment is right on. I only have one Dandelion, on my BR-2A (the rest of my lenses including PN-11 I modified with Bjørn's chips), and it is always a bit tricky to mount an requires a lot of care both on insertion into the mount and twisting it into place. Even if I built up well with marine epoxy at the rear and the sides, images presented at NG before show that the plastic around the contacts on the front is very thin.
I also wonder about the long term durability of the type of plastic used in Dandelions. Certain plastics like the one used for buckles on my backpacks and bike panniers seem to break down with time (possibly due to UV exposure?), and then they start self destructing one after another (often several within a short time span) with almost no load at all.
-
I think this comment is right on. I only have one Dandelion, on my BR-2A (the rest of my lenses including PN-11 I modified with Bjørn's chips), and it is always a bit tricky to mount an requires a lot of care both on insertion into the mount and twisting it into place. Even if I built up well with marine epoxy at the rear and the sides, images presented at NG before show that the plastic around the contacts on the front is very thin.
I also wonder about the long term durability of the type of plastic used in Dandelions. Certain plastics like the one used for buckles on my backpacks and bike panniers seem to break down with time (possibly due to UV exposure?), and then they start self destructing one after another (often several within a short time span) with almost no load at all.
The BR-2A can also be modified with one of Bjørn R's chips and an AF-S contact block.
-
Good to know if the Dandelion breaks, I thought the AFS block would be too deep, but never really checked it.
-
On a related issue (which is what got me to join this forum - someone pointed me here from NikonForums.com); is there anyone making T2 mounts with Nikon AF confirm chips that are actually worth using?
The ones I've seen on eBay look like the chip will break off if you breathe on it. What I want to actually use this for is metering, not AF confirm, with a cheap but reasonably sharp 500mm F8 lens. I've used a round-about route to check that metering ought to work - I also own an Eos body and a focus-confirm M42 adapter, and used the lens on that with an M42 mount. The results were reasonably good for exposure, not wonderful for focus confirmation, but I know that's only supposed to work at f5.6 and below. All of the Nikon mounts on eBay seem to be made or sold by Pixco, but I can't find any reviews or additional information.
Anyone know Pixco, or a good alternative?
-
This is how mine started to deteriorate, then fail completely:
The start of the end is nigh:
(http://members.optushome.com.au/aking83/D800E_DSC_1492.JPG)
fatigue on the leading edge corner.
coroners report:
(http://members.optushome.com.au/aking83/D800E_DSD_1111.JPG)
Plastic corner collapsed inwards, and pin to the far right compressed and lost it's ability to return back out.
Once it broke in such a manner, curiosity got the better of me and I had to carefully prise it open to see it's inner workings.
(ie. the shell didn't break in half like this, just the one corner broke and made it easier to prise apart.
I do remember that in prising the shell apart, the plastic did break into smaller fragments rather than come apart in one 'half' piece.
-
Compare this to the "real thing";
robust contact block with sturdy metal guard, bespoke metal screws for mounting instead of glue, silver contact pins with lateral spring action (thus optimised for the situation of putting the lens to the camera)
Need I say more?
-
Sounds like I should forget about dandelion chips for the Nikon - the Canon ones seem a lot more robust, but of course the spring contacts etc. are in the body, the chip just has flat contacts. Fortunately it isn't that big a deal, I'm reasonably good at guesstimating exposure with the 500mm lens, or I can use it on my Eos instead.
Are you salvaging contact blocks from Proneas? I see them around cheaply occasionally, it looks like the block is the same.
-
No, these are factory parts ordered from Nikon.
The Pronea contact blocks were made all in plastic and had an inferior construction compared to the later version. This extends to the build and material of the silver pins as well.
-
No, these are factory parts ordered from Nikon.
The Pronea contact blocks were made all in plastic and had an inferior construction compared to the later version. This extends to the build and material of the silver pins as well.
I obviously haven't looked at them closely enough. OK, thanks for your help, I think I'll leave this idea unless someone makes a much better chip.
-
No, these are factory parts ordered from Nikon.
The Pronea contact blocks were made all in plastic and had an inferior construction compared to the later version. This extends to the build and material of the silver pins as well.
Bjørn remember to put an all plastic short type contact block aside for the 400mm - 600mm dual CPU CU-1 Focusing mount ;) You probably all ready did I guess ::)
-
Oops did I miss something Erik - you wish to doctor your own CU-1 adapter? Just say and I'll fix. Anything for you my friend.
-
Not mine :) But for your adapter, just in case to avoid a shortcut between the contact pins and the rear plate,,, Not sure how long the thin tape will last.
-
Aha, thanks for the heads-up. I think there are a few of the older type floating around in the scrap bin.
-
Hi,
I'm considering buying a Nikkor 28mm f2 chipped for aprox. 150 Eur. Is it any way to find out if Dandelion chip is used?
-
This is not chipped wit a Dandelion.
This is an original CPU contact block, so a Bjørn CPU - screws for the contact block are sticking out a bit, so check if it mounts the camera ok,,,
-
Thanks Erik, I will have a look at this lens at the end of the week.