Charlie's Shed - Lining a petrol tank

Regular readers will know that in the past few months I have made weld repairs to two petrol tanks, one of which still has a weep in a very inaccessible place.

Today I'm going to line the tank with Slosh, from Rustbuster. This is a polyurethane sealer, resistant to Ethanol, that cures in air. I've cleaned this tank already with their SP10 Tank Kleen and a selection of nuts and bolts:

Those bolts came out a lot cleaner than they went in!

The sealer comes in two bottles, for two coats. You have to pour it in, rotate it around, wait twenty minutes and repeat twice more.

You let it drain from the tap connection to avoid it pooling in the bottom of the tank.

The reason you don't want it pooling is that it gases off a little when setting, and you don't want voids in the lining:

Three hours later, or when the sealer is dry but still tacky, you do the second coat in the same way. Then you let it cure for five days. Easy peasy.

This 400 ml kit is supposed to be sufficient for a 10 gallon tank, and I'm sure it is. My plan was to seal both the MZ tank and the Huntmaster tank in one sitting, but obviously two tanks have a greater metal surface than one tank (though the two together totalled less than 10 gallons) and I wasn't confident I'd have enough for both, so I ordered some more sealer for the FH.

Now that son Tom has his MZ sparking, he's clamouring to get his tank back...

Charlie's Shed - Points backplate

A plea on the VMCC forum found me a little job this week - local BSA B33 rider Richard was looking for a new points backplate - the old one had a stripped hole at the contact set clamping screw.

I volunteered to help - I do this sometimes, it’s a good way to meet new people, it costs nothing and it’s always nice to practice one’s skills. A quick Google revealed that while you can get contact sets, caps and condensers for these ‘distributors’ - Lucas 18D2 or similar, used by BSA, Norton and Royal Enfield the backplate is not available:

The 2BA hole had no thread left and the screw was in a bad way too. I started by marking the hole position on the back, with a scriber:

Laid on the usual copper bar heat sink, I set the TIG welder to 35 A and filled the hole from both sides.

I ground it flat with the Dremel and a file, drilled it for tapping 2 BA (4 mm) and recut the thread.

I cleaned it up with a wire brush. It took a bit of Googling before I was sure I had it assembled correctly:

Having delivered it back to Richard, and seeing his lovely B33 which he has owned for 16 years I was pleased to learn it started straight away. I must have assembled it OK!

It was great to find a few beers on my doorstep the next day. That’s the kind of fee I like.

FH - swinging arm trick

 Ariel PRF frames have a close fitting swinging arm that looks like it comes out backwards.

It doesn't, because the curve of the frame tube gets in the way:

The trick is to push it forward - this obviously only works if the chain guard and oil filter, and the rear mudguard, are already removed. 

FH - it's coming apart!

 I think all this post is going to do is perhaps convince readers that I am actually doing something. It's coming apart for painting. 

FH - Petrol tank dent repair

 I know I know I know... this bike build has been going on far too long, but in my defence I do have a lot of stuff on. Recently, apart from the puncture I have been faffing about with a press stud on the W/NG panniers, had the carburetter float bowl apart (I must fit a finer fuel filter, the bowl is full of debris and it's stopping the float valve seating), welded up an MZ tank etc. etc...

After several weeks of trying to pull a massive dent out with stick-on dent pullers - I event tried soft soldering one on - I decided the only way I was going to get the dent out was to cut it out, bend the material back to shape and weld it back in. Buoyed with confidence from the MZ TS125 tank, which came out with only two pinholes I decided to go forth with the little Dremel discs:

These discs are one of the best things Dremel make, and the panel was soon out:

Using a variety of tools, including home-made bending irons, adjustable spanners and hammers I adjusted the body of the tank and the panel I had cut out to match a template I hade made from the good side:

I've bought a variety of strong magnets over the years and held the panel in place for tacking:

The plan was to minimise distortion by making short 1/2" welds and letting them cool before moving to a different area, building up the full peripheral weld in short, cool bursts. I started with six-seven tacks:

And ended up with a fully welded panel:

This was dressed back with the Clarke power-file - a 1/2" belt sander, which is proving to be a very useful tool. Next, I closed off the filler neck with duct tape and attached a cycle pump to the fuel tap:

Result! Two small pinholes, fixed in seconds:

Next, Slosh sealant - to ensure an expensive paint job isn't going to be ruined with a fuel leak.

W/NG - a puncture

Another unexpected event this week - a puncture. Strange, because air suddenly started hissing out of the tyre after I'd been standing in the workshop for an hour - the bike hadn't been out for days. 

I had a spare tube and resolved to fix it with the tools I had on board, just to prove that I could. 

The wheel came out easily, even with the panniers in place, though one of the top hat washers on the drive side (the repro one) was very stiff in the slot. I skimmed it in the lathe. 

Proving that tyre changing is all about technique, the cover came off easily. 

Unusual puncture - the tube has fretted against an unused security bolt hole in the rim:

It's gone through the rim tape too. 

Nothing much here, but I pulled off this old canvas tape, wire brushed the whole rim and laid a piece of the old tube over the hole to reinforce it. 

Having inspected the brake, which was showing nice even wear, I replaced the tube. 

The cover went on without any drama, though initially I forgot you have to start opposite the valve to avoid nipping the tube. 

Job done - almost. The connector hose on the pump was completely knackered. Luck was on my side, because although I inflated the tyre with my workshop pump I found I had a genuine Apex pump with its long connector in stock!

W/NG - An unexpected event

A few weeks ago I was cracking on home on the W/NG, doing 55-60 up the North Walsham by-pass (the old Midland & Great Northern alignment to Honing) when she suddenly cut out. After finishing the ignition checks I kicked it over to find apparently no compression. I popped the exhaust valve cap off for a look, found nothing and a few minutes later she started OK and carried me home apparently fine.

Stuck valve perhaps? She was pretty hot...

At home, I put the endoscope down the bore for a look and it was clear there had been a partial seizure at some point. Thinking about it, there had been a squeak from the rear tyre and a little shimmy at the back end, and I'd had the clutch in very quickly.

Pulling the head off you  could see the scuff in the bore very clearly:

The rings were OK, apart from the fact that the second ring was tight in it's groove. Hardly surprising really:

Rightly or wrongly, I polished this out. Doubtless someone will tell me I should have replaced the piston.

I built this tool to hold the barrel while I honed it - it's just a couple of bits of plywood:

You can mount it in the vice either way, but this is the most convenient for honing. I mounted it vertically to grind the cylinder head to barrel joint.

I used a cylinder hone to clean up the bore. Hopefully these honing marks will retain a bit more oil...

I also replaced the inlet valve guide which was very loose - the new one had an interference fit in the head but was a smaller bore size - too small for my valve stem. I bored it on the lathe to change the diameter from 5/16" to 3/8", but writing this post I've realised I probably have an exhaust valve in the inlet position... 

Anyhow, it's all gone back together and is running beautifully, apart from a pesky leak from the decompressor. I've reverted to fibre rocker box gaskets (I can't get the aluminium ones to seal) but I've still got a weep from the decompressor, which has a very worn stem. 

It'll take a bit of work to fix that as I want to re-engineer the sealing arrangement, but that will have to wait. It's been OK for the last 70 years. 

Mini-lathe - Another little accessory

A tool height guage! This handy little accessory is made from a bit of 1/2" square bar, polished on the bottom face where it stands on the bed ways. The vertical bit is a piece of strip welded to the base, with a notch at centre height. 

It's used to set up the tool height so it's in the desired position relative to the spindle axis. 

Mini-Lathe - A couple of modifications

It's been a while since I wrote anything about the workshop but recently I made a couple of small modifications to the mini-lathe that I'd like to record in this post. 

First up, a new chip guard made from a 2 mm sheet of perspex, mounted on a steel arm with a strong magnet to sit on top of the lathe:

Secondly, I have modified the sleeve nuts used to retain the splash back - these now comprise a 20 mm spacer and a shorter sleeve nut; the splashback now sits 20 mm further back and is no longer fixed directly to the bed. This means that:

• chips can be pushed into the tray from the splashback and removed from the front
• the cross slide goes to the end of it's travel without hitting the splashback first.

W/NG - Dealing with leaks

 Long fast club runs tend to reveals failings in your bike, especially when 'given the beans' by other riders. You know who you are! The oil leaks that have plagued the W/NG really made themselves felt on the 87 mile run around Norfolk that the VMCC Blakeney Run entailed. It was clear that the oil was coming from the top end, but I wasn't sure if it was the decompressor, the rocker box gaskets or the cylinder head bolts.

Having spent some resealing the decompressor recently, I removed the rocker boxes to take a look revealing that the existing aluminium gaskets did not fit at all well:

 I resolved to make some new gaskets.

I made these from Flexitallic pink gasket sheet, which I think is rated to 350°C. I made the gasket a bit wider than the aluminium one, to make sure it covered the whole joint. I sealed it in place with Threebond 1215.

Preliminary testing suggests this has sorted most of the problem, but the decompressor is still weeping a bit. I'll have to add some more shims.

Charlie's Shed - Speedometer Cable

For some bizarre reason the speedometer cable on the W/NG broke the other day. It's been there for a couple of thousand miles, but wasn't particularly well lubricated so perhaps that was the reason. I'd not considered repairing a speedo cable before, but there wasn't a replacement available commercially and I needed it for the VMCC run to Blakeney the following weekend, so I decided to attempt a repair.

The first attempt was to use the wonderous TIG set to weld a short length of 4 mm round bar to the end. I filed it square and it looked fine:

I made a little collar out of a bit of 8 mm aluminium round bar and used Araldite to hold it in place:

It lasted 27 miles before it broke in the same place.

Looking at the brand new cable in the Huntmaster, I found that the square drive end of the cable was made in aluminium and crimped to the cable end. I made up a similar fitting in the lathe:

Glued in place, with JB Weld this time, I fitted it to the bike.

I'd also bought a cheap Garmin GPS cycle speedometer, thinking that I'd need one if I ever used the Model A on a club run, which I certainly will do at some point. I'm currently testing the new cable against the Garmin, which reveals that the Chronometric on the W/NG reads about 5% high.

SQ4 Clutch Investigation

 After the Blakeney run, Tom reported that the SQ4 clutch was slipping when he got home - he couldn't get it to go over 50 mph without the clutch slipping. 

We rebuilt it using the new plates from the W/NG, which got it's old Ferodo plates back; middle grandson Felix helped with the job:

We made a little pointer to show the pressure plate was lifting squarely:

Oddly, it wouldn't function with the old pushrod - We had to add 1/8" to the pushrod length, having found the original 12 3/4" wasn't long enough but 13" was too long. We settled on 12 7/8".

A month later, we had the opposite problem where it wouldn't disengage - Tom took it apart again and adjusted it, and all was well. Sticky new plates, or some intermittent fault?