Nikon lens repair: AF-S DX 18-55 f/3.5-5.6 VR

Here’s the background story: my Father-in-law had his Nikon D60, with 18-55 VR kit lens attached, on his shoulder when he dropped the lens cap. The cap rolled away and was about to disappear out of reach, so he ran to get it. Unfortunately his Nikon slammed against a metal railing, lens first; the lens came away, would not re-attach and there was no knowing how much damage to lens and camera had been caused.

When it came to me I could see that the whole mount was badly warped and that bits of it had broken off. The electronic connector block was hanging out and the surface which is supposed be flat to mate with the camera throat looked like a tiny roller-coaster track. My F-i-l was going to dump the lens but what he really wanted me to check was the camera, to see if it had been damaged by the impact. My first thought, though, was to save the lens from going to landfill.

It helps that I have taken a few of my own Nikkors apart before, for minor servicing/repair, and so I went straight to remove the whole mount assembly. I found the cause of the severe warp: a tiny lug had broken off inside and was lodged underneath the shims that it had previously been holding in place. I determined that I could get by without this lug, since it was too small for a reliable fix and the shims had one remaining lug to hold them in place before the main mount screws secured them. I re-assembled everything as best I could (the electronic connector block was still a bit loose as one of the screw housings had also broken away) and tested the lens. I had to hold it against the camera body, since two of three mount catches were missing, whilst testing AF, aperture and VR. All seemed to work fine – which was good news because that meant the camera was probably OK too. I feel sure that, had the lens mount been the usual metal instead of plastic, there would have been some significant damage or mis-alignment inside the camera. As it was, the mount absorbed the impact, breaking and warping, thereby saving the camera from any serious damage.

OK, so everything worked, but I wasn’t sure where exactly to get a new mount. I looked for similar lenses on a well-known auction site, being sold for parts, but found none. Thankfully I found a company in the U.S. which sells parts for camera repairs and they had the correct mount for about a tenner. So I ordered it and it arrived a couple of days ago.

The repair itself is quite straightforward. The three main screws are removed, as well as three much smaller screws holding the central sleeve in place. There are two more tiny (but different) screws which hold the electronic connector block in place. There will be a number of shims, which will be specific to each lens and they are to adjust the distance between main mount and optics (i.e., sensor to lens distance). The image below shows the lens barrel, the broken mount to the left, the electronic connector block and, in the background, the inner mount sleeve and shims. Click on any image for a larger version.

Two things need to be done before the main mount can be replaced: the sprung aperture guide is removed to be attached to the new mount as is the small connector with sprung pin, shown near the centre of the image above. Now we’re ready to prepare the new mount.

The aperture guide slots into the underside of the mount and the spring is attached:

The sprung pin is attached:

Then the shims are returned. Notice the guide lug at the top right; there should be another on the bottom left next to a screw hole but it’s missing, so the shims can shift around a bit. This gets fiddly when finally putting it all together as the aperture guide needs to be carefully placed on the correct side of the aperture pin in the lens, otherwise the aperture won’t work at all, but getting the correct positioning can move the shims so that screw holes don’t line up.

With the aperture guide in the correct position and the main screws loosely in place, all that remains is for the inner sleeve to return to its rightful place. Care must be taken with the five tiny screws: two of them have slightly larger heads and a finer thread pitch. It is these two which hold the connector block in place, whereas the other three secure the inner sleeve to the main mount.

After tightening all screws (not too tight – this is a plastic mount, after all), a good clean and mount to camera body, all is well. Focus is accurate and VR works as it should. One more saved Nikon lens to add to my count (OK, that makes two in all – but hey).

One final thought: those shims are thin. Some of them are extremely thin. They are of different thicknesses and will have been put together in that combination for this particular lens and its original mount, so it could be argued that they are no longer an accurate set for this lens. Since all my testing shows focus to be spot-on, though, they appear to be good enough.

Hocus Pocus with vinyl records

Every so often I’m treated to a stack of vinyl which I hope will contain all those taped records I never actually owned back in the ’80s and early ’90s, as well as other interesting nuggets from the ’70s and beyond. Yesterday was one of those days, when I discovered that the nearest decent town does, in fact, have an independent record shop, tucked away in behind an antique shop in a part of the town I don’t normally see.

The problem with used vinyl is, of course, grime and scratches. Nothing much can be done about scratches and dents; visual inspection before buying is a necessity but even that won’t always show up problems within the grooves and so, to a certain extent, it’s a gamble. Dust and grime, though, needs to be carefully removed for obvious reasons but also to minimise further wear. There are expensive solutions to record cleaning with purpose-built machines costing hundreds (there are one or two “cheaper” ones but I’m not convinced of their effectiveness) but I thought I’d come up with my own method, which works very well. Records where the grime in between the grooves would not move with any amount of carbon-fibre brushing and careful wiping are now as good as new.

What I use: two microfibre cloths (it’s important that these are soft microfibre – I use Vileda ones), washing-up liquid, Ilford photographic wetting agent (I have loads of this as I process film), 99.8% (or better) isopropanol and soft cotton pads (which I steal from my wife’s stash).

The record first has a wash in warm water containing a squirt of the detergent and a splash of wetting agent. It’s not immersed but held carefully whilst gently wiped around with plenty of fluid washing over it, using one of the microfibre cloths. After rinsing with clean water it’s dried on both sides with the 2nd microfibre cloth. Finally, in order to get rid of any remaining stubborn marks and water mineral residues from the water rinse, careful wiping with a cotton pad generously moistened with the isopropanol completes the clean.

The result is no more grime, minimal dust, little surface noise and, unless the record had been previously damaged by bad handling or a worn stylus, fantastic sound! It’s taken me a few goes to get the process right but I’m very happy with the results.

I’m currently listening to 90125 by Yes, one of the four I bought yesterday. It was the worst of the lot as far as dust and dirt go – a real state – but, as far as I could tell in the shop, no real physical damage. I’m surprised at how well it cleaned up and at how it sounds now. Pretty much as good as new. I’m back in the 6th Form common room with this blaring away, although it obviously sounds much better now

Another was Greatest Hits of Focus. I’d been looking for a decent copy of Moving Waves only really because I wanted the track Hocus Pocus; a song I’d heard when I was about 18 but had forgotten about until an advert on TV recently used a snip from it. Yesterday I found a copy of Moving Waves but it was clearly damaged and they wanted £4 for it. Next to it was a Greatest Hits in excellent condition for £3.50 and so, whereas I normally avoid hits albums, I took it because the frankly brilliant Hocus Pocus was there. The album actually contains a number of tracks that take me back to the 1970s and I didn’t even realise they were Focus, including House of the King which I’d thought was Jethro Tull!

The other two? Cyclone by Tangerine Dream and Eliminator by ZZ Top. Spin those guitars!

Creek Wyndsor

CD loudness war

I’m getting really hacked off with CD “brick walling” (Google “CD loudness war” to see the subject of my latest rant). I’ve recently bought a few CDs and the majority are affected by this stupid obsession with making the sound as loud as possible, at the expense of the music itself. What I’m hearing is the potential of good music which is veiled by clipping, distortion, crackling and generally being so loud that everything is at the same volume. There’s practically no dynamic range, at all, with everything being pushed up into the last few available decibels. The result is that waveform peaks get chopped off – so effectively there are no longer any peaks but instead there are “flats”, leading to lost information and distortion. A comparison between your typical, recent CD and a recording released now by B&W’s Society of Sound illustrates the glaring deficiencies of many recent commercial releases. What a difference. Generally speaking, my early CDs sound so much better than later ones, and I’m not talking about the style of music.

It’s getting worse. These days, a “digital remaster” re-release of an older album generally means that it’s louder. So that’s good, isn’t it? Well, no. Record companies need to understand that we’re not all earbud-toting, cloth-eared teenagers and give their mastering engineers some room for manoeuvre. Louder doesn’t equal better, for goodness’ sake. I would rather turn up the volume of something I like, than turn down the volume because it’s a wall of noise.

It’s no wonder the popularity of vinyl is on the increase.

Windows 7 “XP Mode” finally useful

I’m partly writing this for my own benefit, as I know I’ll forget this at some point and need to re-set it…

OK, so Windows 7 is what Vista should have been, blah blah… I actually quite liked Vista Ultimate x64 but I’ve gained some speed and UI niceties by changing to W7 Ultimate x64. In both, I’ve been using a modified driver for my Nikon CoolScan 5000 which Nikon does not support because the company refuses to release a 64-bit driver for their film scanners. They have effectively given up on film scanning now (the 9000 is the only one currently available but it’s too expensive and probably won’t be around for much longer either) which is a pain because their scanners are good and there’s not a lot else to choose from these days. 32-bit Windows: no problem. 64-bit: forget it? Not quite.

So I figured out how to use a driver on x64 which would drive the 5000 (I love the internet) and it does work. Usually it works well but I’m finding it occasionally crashes, especially if I’m using the post-processing options such as GEM.

One big selling point of Win7 Ultimate (and Business/Enterprise) is their “XP Mode”. This is basically a registered copy of XP SP3 running as a virtual machine under Win7. It allows the connection of USB devices and, even though the VM is running on Win7 x64 in my case, the XP within is 32-bit. Great – I can run the supported scanner driver under XP 32-bit under Win7 64-bit! Damn – the VM is limited to 16-bit colour and, therefore, looks crap.

So I gave up on XP Mode and put up with the occasional crashes until a few days ago, when I came across this:

http://www.mydigitallife.info/2009/12/16/how-to-increase-color-depth-and-quality-of-windows-xp-accessed-via-remote-desktop-or-terminal-services/

I knew I could increase the colour depth if I disabled the integration facilities but that would mean no connection of USB devices to the VM, which would be pretty pointless. This link shows how to get around the 16-bit colour limitation to give 24-bit colour whilst retaining integration. It turns out that XP Mode uses the Remote Desktop Protocol (RDP) and this is limited by default to 16-bit colour – presumably in the interests of connection performance. But I’m not connecting to another physical machine here. This link shows that a simple Group Policy edit (GPEdit.msc) to Terminal Services to set the maximum colour depth to 24-bit should do the trick and it certainly does. Suddenly, XP Mode is useful and I can scan within it now with everything looking as it should.

Milestones

Mr Monkey is eight months old today – and celebrated by standing up for the first time in his cot! Last weekend he’d started crawling, just in time for a visit from Granny and Grandpa who hadn’t seen him since Christmas, so the rolling (his previous method of transportation) was relegated to changing direction. He’s been trying to pull himself up on things ever since and this morning he surprised us both by standing up and looking very pleased with himself.

It’s fascinating to watch as he develops new skills and all in a seemingly short amount of time, and it seems odd how easily we forget what he used to be like. I’m trying my best to capture as much on film and sensor as I can and have created two DVDs so far; watching them as we did recently, we couldn’t believe how small and vulnerable he was in the early days – unable to co-ordinate and barely showing recognition. How things have changed. He’s like one of those copper-top battery toys on TV that just go on and on until the batteries run out. Except they don’t.

Here’s a snap from this morning – caught red-handed opening the kitchen drawer:

DAK Novo pt. 3

There are optional tweaks mentioned for the Novo and I decided to try a couple of them out. The first, fitting optional twin capacitors, fell at the first hurdle (buying the capacitors – only one in stock!) and so it was time to thermally couple some transistors (oh yes).

I shan’t pretend I really understand why, but it’s theoretically beneficial to have a consistent temperature across each of two pairs of closely facing transistors on the board. In each pair, though, one transistor is likely to get a bit warmer than the other. Having them as closely facing as they are is probably enough but my obsessive nature meant that I was always wondering whether I could/should tie them together somehow. It’s been mentioned that an elastic sleeve should do the trick but I decided to go further: heat-transfer grease and heat-shrink tubing.

With a cocktail stick I applied the special grease to the facing surfaces of the transistors (quite a bit of prodding was necessary as they were close together) until I was satisfied that there was enough between them. This grease is the same stuff as is used between CPUs / GPUs and their heatsinks so the idea is that the warmer transistor transfers heat more efficiently to the cooler transistor in much the same way that a CPU dissipates heat through its heatsink.

I cut some heat-shrink tubing to give two sleeves which were placed over and around each facing pair. The hard part was moving the tip of a hot soldering iron around the sleeve several times, so that it shrank, but without touching anything. Not a job to do with a hangover! I then put a little more grease on top, around the join.

That’s it. It’s questionable whether this will make an audible difference – it’s supposed to but I guess it’s clouded by the fact that my Novo is still burning-in, having been powered for a little over two weeks.

DAK Novo pt. 2

Next in were the rest of the capacitors, followed by the input and output jacks. The potentiometer supplied to take care of volume control needed to be hacked about a bit: first to remove a lug and then to cut the shaft down to size – this was probably the one time I was worried I’d trash something. Thankfully, I didn’t. I started off with a hacksaw and then decided a Dremel with rotary cutter would do the job better. With that finally soldered into place, there were a few odds and ends to connect to the board, including all earths, and the LED.

Almost done...

I powered it up. The green LED lit. Get in!

Nothing blew up or started smoking, so I measured the voltages across several components as described in the booklet. All seemed well. Once mounted in the case (quite straightforward) it was time for the ultimate test: sound.

Case earths connected

It works! And not only that, it sounds good! With a source unit (CD, MD, whatever) set to line-out instead of headphone-out and a decent set of cans, such as the AKG 702 which I happen to have, the sound is noticeably better in every respect compared with the built-in headphone stage of the source unit. The Novo is used in this way instead of the source unit’s own headphone amp or it may be used to provide a headphone out for a system that doesn’t have one. Especially in kit form, this is a bargain for what it does; yes, it was a few hours’ work but there’s the satisfaction of building it thrown in for free. Good stuff!

Listening test

DAK Novo headphone amplifier kit

So I decided to get a dedicated headphone amplifier for the AKGs even though they sound pretty amazing with my portable units. It’s fairly well known that the amplifier stages of most portable units – and also those built in to most full-size separates – are compromises in terms of cost and available size and so I went looking for dedicated amps. They naturally varied in cost from a few tens of pounds to literally thousands and I had to set a limit as to how much I was prepared to spend. One name kept cropping up in the more affordable range: Graham Slee Projects. Graham and his team are U.K.-based audio engineers whose products – mainly headphone amps and phono stages – have been getting rave reviews for some time. Their Novo Discrete Solid State Headphone Amplifier has won awards from What Hi-Fi and can be bought direct for a very reasonable price, but it’s also possible to save a bit of money and buy this amp in kit form. The saving is around £80 and so it’s worth it, but only for those who are confident with electronics and soldering. I’ve messed around with electrics/electronics and soldering on and off since I was fairly young but I’ve never built anything like this from discrete components, from scratch. This was a challenge I had to take.

When bought in complete kit form you get exactly the same components, exactly the same case (except for a couple of cosmetic differences) and exactly the same power supply as for the retail version. The cosmetic differences are visible on the front of the case: the name and font are different (to reflect that this is a DAK – DIY Audio Kits – amplifier) and the front screws are different. That’s it – otherwise exactly the same.

The first thing I needed to do was identify the components, particularly all the resistors. The supplied booklet helps a great deal here, reminding me how to read the colour codes (although I don’t think I’d ever come across a six-band one before so I looked it up just to double check that the bands meant what I thought they did) and a multimeter helps to verify if necessary. I suppose I could have just used the multimeter to read them all off in the first place, but where would be the fun in that? A cross-reference with kit numbers enabled me to identify each one in relation to the board.

Resistors identified and marked

Component kit with my adjustable grip holding the PCB

The next step was to insert the resistors and solder them into place. I’d bought some 60/40 tin/lead 0.7mm solder especially for this and it does the job nicely. I’d say my soldering skills are fairly basic but there’s nothing difficult here, given some care and attention.

After snipping off the excess pins underneath, the diodes went in next, followed by ceramic capacitors and transistors. With the bottom of the board becoming more and more busy with soldered joints, some concentration was needed for the transistors – especially as the pins should be soldered in a certain order.

Resistors and diodes in place

Transistors are in

Underneath

To be continued…

Weaning

First solids

Yesterday, at a few days under six months old, Samuel had his first taste of solid food. Of course, I was there with camcorder in one hand and camera in the other; I managed to capture the event and still enjoy it as it was quite funny. Not sure quite what was going on at first, it didn’t take long for the spoon to be grabbed and the food guzzled down! He was certainly ready for it, that’s for sure.

The brunch menu consisted of baby rice. Samuel chose baby rice from the menu and I believe he’ll choose the same again today. The menu will soon be expanded to include things like broccoli, carrots, parsnips and such like, so there are exciting (and very messy) times ahead.

AKG K702 headphones

My old Sennheiser HD495 cans were starting to show their age – mainly through the earpads’ and head pad’s disintegration – so I decided it was time to semi-retire them and get something new. I also wanted to take the opportunity to replace them with some higher class headphones and so my search began, albeit without the opportunity to actually hear any.

Before long I had shortlisted a few and kept coming back to the K702, being described as revealing, flat-response reference class headphones. Basically, the descriptions of the sound appealed to me even though some said that the sound was a bit uninvolving (I ignored any comments from obvious “bassheads” who raved about their cans and put these down as I’ve heard the type of overly warm, overblown bass that they like and it’s not what I want). Although the K702 is aimed at studio recording/mixing use, its sibling, the K701, is aimed more at audiophiles and the sound is, apparently, practically the same.

It’s said that the K702 needs around 1000 hours’ running-in time before they really shine. Some say that the sound has settled nicely by about 200 hours. My initial impression was disappointment as the sound seemed distant and light.

Approximately one hour later my opinion had reversed totally and I could hardly believe  what I was hearing. What I hadn’t taken into account, clearly, was that I needed a little time to get used to this sound; all of my music sounded – sounds – different, hence the initial disappointment. It’s night and day. My 595s sound as though they are being pushed into my ears via a pile of old socks, in comparison. There’s just no real detail there but the K702, by contrast, reveals everything – both the good and the bad of any source. On balance this is undoubtedly good since I want to hear what’s there, rather than some veiled “best effort”. Having listened mainly to compressed music – usually no lower than 256 kbps ATRAC or AAC –  I can say that these ‘phones are a definite improvement at every level compared with anything I’ve used before and I’ll say the same for the few 132 kbps ATRAC tracks that I have.

I’m hearing things I never knew were there and everything has opened up considerably. If they are this good virtually out of the box, then I am very happy indeed as I’m having trouble imagining how they could improve with running-in. But that’s what everyone who owns these seems to say, so we’ll see what happens. Incidentally, it’s often said that a decent, dedicated headphone amplifier is pretty much a requirement for these in order to get the best out of them. That I can understand, especially since their sensitivity is lower than all my other ear/head phones, but so far I’ve only used the K702 with portable music systems (iPod, minidisc, etc.).

If there’s one niggle then it’s this: when I first put them on and for a few minutes afterwards, the pads seem to put a little too much pressure on the areas directly below my ears. It starts to feel uncomfortable so I need to pull them away a few times until I get used to them. No big deal, and it doesn’t seem as bad after a few wears as it did at first.

Other than that, I am extremely happy with them and, if they improve with running-in and/or a headphone amp, then happy days.