Category Archives: Panther

Panther motorcycle restoration

Lights On – Again

Now that I’ve taken the sidecar off the Panther I can again ride her as a solo.

First time out felt a bit strange, probably because it’s been on sidecar for a while and the tyres will have worn to suit a sidecar outfit.

Since it’s only been for around 1500 miles they’ll soon wear back to suit a solo so that’s no real problem, especially as she will not be used as a performance machine.

Once I got back home I realised that the headlamp was not working, in fact none of the electrics were!.

First thoughts were that I’d managed to blow the main fuse but a quick look showed that this was not so, and then I saw that the stop light was still working.

This showed that the battery was still charged and its connections were sound.

But the stop-light is not fed from the headlamp area, it has its own feed from the battery, so suspicion fell onto the wiring and it was out with the multimeter.

The wiring system on this old lady is laughably simple when compared with that on a modern machine.

There is a wire runs between the battery and the ammeter.

Another wire runs between the regulator and the main switch.

One runs from the switch back to feed the rear light and then I’ve added an extra one from the headlamp shell to the frame ground point and that’s the main fore and aft loom!.

There is in addition a connection between the ammeter and the main switch and then there’s the wiring going to the headlamp itself plus that to the speedometer bulbs and to the horn.

The headlamp is normally grounded through the frame of the bike but since I had added a grounding wire as a back-up, suspicion was on the battery lead to ammeter wire.

I used the multimeter to test between the ammeter and the headlamp shell and sure enough there was no power showing.

Suspect point was where the loom flexes as it exits from under the fuel tank and crosses to the headlamp shell.

Normally there would be only a small amount of movement here but I’ve had a sidecar fitted, this means that as with a sidecar fitted you actually steer by turning the bars there is much more movement, and hence flexing of the loom here.

Sure enough, I could feel a break in one of the wires here and so decided to splice in a new section of wire.

This meant taking the tank off to get access, so it had to be drained!, typically I’d just tanked up, luckily only to half-full!.

I opened up the loom at the break to find the wire had totally parted and so I pulled that wire out of the loom further back to under the tank to make a splice.

Until now I’d assumed it was the main battery feed wire forward had parted but it turned out to be the additional grounding wire I’d added! Shows how unreliable a frame ground can be!.

To be safe I checked by running a temporary ground wire from the headlamp to battery ground and – Lo and Behold everything worked!.

This made life much easier as the ground point inside the headlamp is much simpler to reach than the power point!.

Rather than just bridging the break I ran the new wire from the splice point under the tank direct to the headlamp ground, this meant I only had the one splice in the wire, less potential for problems!.

A swift check of the system again and everything was working so it was just putting things back together, add fuel and we were back on the road.

What’s Happening??


It’s been a while since I posted anything on this blog, too long! But there’s not been much happening with the bikes as they’ve been running well. However I’m feeling in need of a change now so–.

I took the sidecar off the big Panther a couple of days ago, not the easiest of jobs when working on your own, but do-able, the body is now up on a pair of trestles in the shed and the chassis is in the back yard.

The Panther now stands as a solo in the garage, but she’s not on her own as the old BMW has been brought out of storage and is standing beside her. They make a good pairing, the 600cc 1937 Panther M100 and the 750cc 1940 BMW R12.

Although the Panther has the smaller displacement engine of the two she is the more powerful at 26bhp to the BMW’s 20bhp, but the BMW is a twin cylinder side-valve with a heavy external flywheel so is possibly the better slogger of the two, still neither was intended as a sports bike.

Anyway the BMW is set up with a full set of Steib quick release sidecar fittings so once I have the chair set up for her I can drop it on and off in around five minutes work, Useful!!

Now, what does the BMW need doing to get her in commission again?

Well her battery is an AGM sealed unit and has survived her laying up, still showing about half-charged when I put it on the charger. She is magneto ignition and that’s sparking well but I’ll treat her to a new pair of plugs.

How about oils?, well oil is cheaper than metal so it’s new oil all round, and owing to her age it’s an old-fashioned non-detergent type oil. Equally I needed new drain and filler plug gaskets.

On a machine of this age these should be the hollow rolled-copper type rather than the solid aluminium ones BMW now supply. A quick rake round on ebay soon found a supplier of these in the neccessary 14mm and 18mm sizes.

Rather than run her up to hot on the old oil I let her drain out overnight and then it was just a case of fill the engine to between marks on the dip-stick and the gearbox and final drive until the oil reached the bottom of the filler hole threads.

Now came the acid test, she’d been stored with a dry tank so I put a half-gallon of fuel into the tank, no leaks visible! GOOD!!, turn on the tap (which way is “ON” and which is “RESERVE”??, I can’t remember!! ).

OK, the carbs had filled so a good tickle on each carb, crack open the throttle and turn over the engine several times to prime the cylinders, then it was just switch on the ignition and kick her over.

While she fired she didn’t pick up so a tad more throttle, try again and the old girl was running again! Tick-over balance is a bit off but I’ll need her warmed through to set that up so it will have to wait.

Now I know she’s a runner what else needs doing.

An immediate obvious is indicators!. Last time out she was hauling the chair.

I know it’s non-original but on modern roads an outfit NEEDS indicators and now you can use LED units without overloading the electrical system so she was fitted up with lights on one side, the others being on the chair. So it was get another pair of indicators and fit them.

I had to make up (and paint) a set of mounting brackets for them, run in the wiring and then tap it into the circuit to the sidecar indicators, a quick check and found they were working.

Next thing was cosmetics, the front mudguard needed some serious touch-up where the paint had been badly scarred and had started lifting while the old lady had been sitting in the storage unit.

Fortunately it can be redone while still on the bike as it would be serious hassle to remove it and then given a dose of “T-Cut” after about a fortnight’s curing followed by a polish job should see things OK.

As I’m writing this she’s standing in the garage with the first top coat on the guard giving it time to harden off before she’s put away for the night, another coat tomorrow should then see it ok.

I can’t really complain as both guards were in a bad state when I got the bike back at the beginning of the Eighties and they needed serious patching to be made usable, since then she has seen some serious mileage!

A Little Light

With the sidecar fitted and aligned, before I can take it out on the road it needs lights to be legal.

With it being a 1937 machine I just need to show “a white light to the front and a red light to the rear”, after all at that time gas lighting was still common, but as this outfit is to be used on modern roads I need to be a  bit more practical.

What I am fitting is a spotlight set up as a Daylight Riding Light, a white front running light, a red rear running light, a stop light and indicators.

Indicators are totally out of period but in modern traffic on a sidecar outfit they are “A DAMN Good Idea!”.

I had a pair of the Hella round indicator/running lights in stock so these became the basis of my lights.

As the wheel and mudguard (fender) are on the outside edge of the outfit the lights need to mount onto this but as the guard is semi-circular this gives a problem, the Hella units are intended to fit onto a flat, vertical surface and the guard has a compound curve, at the point where the lights need to mount there is a 45° slope to the vertical.

First thing was to make the mounts for the lights. These need to fit onto the guard and give a suitable surface to mount the lights.

As the lights are 75mm in diameter I took a 6 inch length of 75mm OD alloy tube with a 1.5mm wall thickness and cut it into two lengths on a 45°angle.

Mount tube

The sawcut alloy tube

These, when mounted onto the guard will fit onto the guard part way up it and give the necessary vertical surface, unfortunately though the plain cut end does not match the curves of the guard.

tube onguard

Showing mismatch between tube and mudguard, it only touches at the ends

However from fitting the guard onto the sidecar I had a short length of it spare

spare guard

“Spare” length of mudguard

so I took this, fixed some 80 grit abrasive sheet to it.

I next took a black marker, used it to “black up” the cut end of the  tube

tube end blacked

“Blacked-up” end of tube

and then started rubbing the mount on the abrasive sheet.

rubbing in

Rubbing in end of tube to match mudguard

This gave me “witness marks” showing where the guard was contacting the mount and so, where the mount needed trimming back.

witness marks

Rubbed end of tube showing “witness marks” where material still has to be removed

This was done using a Dremel tool and a sanding drum.

Repeated re-blacking, trials and trimming gave me a pair of mounts that were a reasonable fit onto the guard and which, given some rubber beading, would make a sound joint against it.

finished job

Tube matched to contour of mudguard

I now needed bases for the lights themselves. To make these I took a pair of 10mm thick alloy disks of about 85mm OD.

These were each chucked up in the lathe

raw disk

Raw disk in lathe chuck

and had a central 8mm hole drilled through.

After this they were removed from the chuck their place was taken by a short length of 32mm OD bar. This had its end faced flat and then had a central hole drilled into it. This was tapped to take an 8mm bolt.


Disk centre drilled 8mm

This gave me a mandrel to work the disk on and so one of the disks was bolted to it, this meant that I could now turn the disk about it’s centre and it was able to have the edge turned down to size.

on mandrel

Disk mounted on mandrel for turning

This disk was now skimmed down to 75mm diameter, the same as the OD of the tube.

Next step was to turn a 4mm deep spigot on the disk to make a tight fit in the tube.

I then reversed the disk on the mandrel and another 4mm deep spigot turned on the other side, this time sized to mount the Hella lens onto.

finished turned disk

Disk turned to size and lens fitted

Suitable mounting holes were drilled and tapped into the alloy to take the lens mounting screws.

All that was now needed was to supply light to the lenses and rather than use incandescent  bulbs I opted to use LED lights instead.

Going onto Ebay, I ordered up 4 amber, two red and two white LED marker lights. These mount with a 10mm stud on their backs.

As these are the equivalent of a 10 watt bulb the intention was to use two amber LEDs to supply each indicator,one white for the front running light and two reds for the stop light.

Going to the disks, I marked off the horizontal diameter through the lens mounting screws and drilled a pair of 10mm clearance holes above it. One one disk I then drilled a single 10mm clearance hole below the centre line and on the other a pair below the line.

drillee disk + LEDs

Front unit disk fitted with LEDs

Mounting two of the amber LED’s above the line gave me my indicators and the fitting the whites or reds below gave me my front running light and my stop light.

I needed a rear running light as well and so a similar but smaller unit was made up using a 50mm diameter LED rear marker light to supply the lens.

All that was left to do was give the mounts a lick of paint, mount them onto the guard via the central hole in the disks and to run in the wiring.

Finished job

The completed light unit in place


Lining it up

It’s now time to start fitting the sidecar. I’ve not yet refitted the tank after the dynamo belt drive conversion so now is an ideal time to do this.

First thing is to get the chassis connected up to the bikes frame.

The chassis was laid out alongside the bike and put up on blocks.

The front swan-neck was put in position on the bike and loosely clamped to the sidecar chassis, as was the rear ball joint fitting.

Fitting a sidecar has been described as a black art, well now starts the black magic!

First thing to set is the lead, the sidecar wheel has to be set a bit in front of the bikes rear one, how much depends on the type of sidecar, the intended use and the bike itself.

A bike with rear suspension needs more lead than a rigid bike; with a heavy sidecar you needs less lead, with a light sports chair you need more. This all boils down to the answer to “How Much?”  being “It Depends” but it is not critical to a fraction of an inch, I’m starting with 6 inches of lead on a light sports sidecar and this can be adjusted, if needs be, after road test. On my later outfit, which does have rear springing, I’m running with 10 inches lead on a similar sidecar.

So the chassis was slid around on the blocks till I had the appropriate lead and the fitting clamps tightened a bit.

setting up

Lead and level have been set, now it’s the toe-in to do. Notice the two “fine alignment tools” by the front wheel

The chassis was now able to have the support blocks removed and, as the bike was now standing on her wheels as well, it was time to level the chassis.

Working with a rigid bike here it was set level side to side, when the bike has rear suspension you’ll need to have the bike loaded when setting this.

While you want the sidecar level side to side, going fore and aft you want it nose high, even when loaded, so the chassis needs setting with the nose rail between a half inch and an inch higher than the axle rail. Getting these right can be a bit fiddly as with the chassis I am using, adjusting the side to side level will also alter the lead if your not on top of the job.

Now comes the most important of the settings, the toe-in.

With a sidecar outfit, the power is all on the bike side, the sidecar wheel is unpowered.

This means that when running straight the sidecar wheel always drags a bit and tends to pull the bike towards it. It is not a heavy pull but it does get tiring compensating for it all the time, so the trick is to set the sidecar wheel to steer a little the other way by pointing it in towards the bike, to “toe it in”.

Thing is “By how much?” and the answer is “It all depends!”. This is the big variable and depends on the bike, the roads you use, road camber will affect it, how fast you are going etc. so you have to compromise and then make adjustments to suit.

Experience leads me to use an initial setting of around 3/4 of an inch over the length of the bike so a straight edge (here a length of 4 x 2 timber) is laid against the bikes wheels, I have the same size tyres fitted fore and aft so it’s set up on blocks and adjusted so as to touch evenly at four points across the wheels.

Similarly another straight edge is laid against the sidecar wheel, touching evenly against it and the distance between the two straight edges measured, first just behind the rear wheel and then just in front of the front one, the difference between the measurements giving the amount of toe-in.

This is adjusted to suit at the chassis clamps, the sidecar wheel being lifted off the ground and replaced and the straight edges reset before a check measurement is taken so as to remove any tyre distortion from the figure.

Once I had the required toe-in the two main sidecar clamps could be tightened up but this is another setting that will probably be modified after road test.

This left me with the sidecar lean out to set. The bike needs to lean a little way out from the sidecar. With a left-hand sidecar in UK you want the bike to be a little past vertical to allow for the road camber, so you drop a plumb line from the handlebar end and set the lean to around a half inch.

setting lean-out

To set the lean-out drop a plumb line from the handlebar.

I’ve got a right-hand sidecar however and any lean I set will be added to by the camber, if I set too much lean I’m going to feel the bike leaning over, so initially I’m setting so as to have the bike vertical on a level surface so on the road the camber will be giving the lean out.

This is set using the rear upper brace, that one goes to just below the saddle.

setting lean-out

Lean -out is set using the rear brace that goes to under the seat

That’s the main adjustments made and locked by tightening the chassis clamps. Some people will tell you these three fittings are all you need but a fourth one is definitely desirable, without it you can feel the outfit flexing in corners, not only that but the toe-in varies as the outfit flexes, Not Ideal!.

This fitting goes across between the front of the sidecar chassis and a low point on the bike to triangulate the swan neck. Normally you would take it off the front engine plates but a Panther does not have these.

However P&M supplied a mounting point on the engine right beside where the front engine supports are so its taken to there. All you do is adjust the fitting so that the clevis bolts slide easily into place and then lock it up in that position.

adding the body

The lower front brace goes between the front of the chassis and the bike.
Once it is in place then the body can be fitted.

With the basic settings made it now requires a road test to determine what adjustments need to be made to these.

What to expect is that as you pull away there will be a small degree of low speed steering wobble that disappears as soon as you are moving. This is normal and you soon do not even notice it, you don’t after all on your car and it does the same. You can reduce this with a steering damper, but too heavy a damper makes for heavy steering, careful attention to the setting up will minimise it, a better remedy.

Now try some slow turns, towards and away from the sidecar, ideally the effort should be about the same but if turning towards the chair is noticeably heavy then you need to reduce the lead to stop the wheel crabbing. But too little lead affects stability and makes the sidecar wheel more prone to lifting.

Does the outfit run straight under steady power? The ideal is to be able to hold it steady with one finger on the bars, does it pull to one side or the other? This is a pointer to the toe-in setting, pulling all the time towards the sidecar — increase the toe-in, pulling away from it — decrease the toe-in.

Open the gas and as you speed up you should pull a little round towards the sidecar; now shut the gas,as the bike slows on engine braking you should feel the sidecar pull round towards the bike.

If at steady speed you can feel it is pulling only slightly but enough to need constant input to stay straight then a SMALL adjustment to the lean-out can help but don’t overdo it, look on this as a final fine trim to getting the toe-in right.

So far all your tests should have been below 30mph, now it’s time to speed up a bit.

Take it gently and try at higher speeds, it may need a bit more fine tuning to get the outfit handling “Just So” but it’s well worth the effort as with a well set up outfit out on the open road it steers largely on the throttle, needing very little input to the bars.

Complete outfit

Final Result, ready for the road

Timing It.

Having removed the magneto and with replacing the drive dog on it with the POC’s uprated version I needed to retime the ignition.

On a Panther there are no external timing marks and the ignition point is normally quoted as with the magneto set on full advance at a set distance before TDC,‭ ‬measured with a rod down the spark plug hole in the head.

An option is to set it to TDC with the magneto on full retard,‭ ‬determined the same way.‭ ‬This sounds easy but it is in fact not so simple or accurate to determine because there is an appreciable amount of crankshaft rotation at TDC with minimal piston motion and it is easy to be a fair bit out.

I decided to make up a special tool for this.

The obvious way is just to take an old plug,‭ ‬smash out the ceramic and use a rod going down through it.

Downside with this is that the rod is going into the cylinder at an angle so that any distance measured on it will not be an accurate measure of piston movement,‭ ‬not only that but because it is at an angle it can jam the piston rather than being pushed up by it and cause damage.

I’ve a spare cylinder head and using this I found that not only could an old spoke go down the plug hole vertically when angled across it but there was a flat horizontal face on the outside of the head at the plug hole to set the tool on and also act as an index point.

Spare head

Head showing flat area at plug hole

So I took a piece of‭ ‬2.0mm flat aluminium sheet scrap and cut it to fit across this flat area.

I then drilled a‭ ‬5/16inch‭ (‬8mm for the metrically inclined‭) ‬hole in the plate,‭ ‬centred on the plug hole and filed the bottom edge of the plate below that to a chisel edge.

Alloy plate

Alloy plate

As I wanted to keep the spoke vertical I scribed a line up through the hole centre and using the corner of a hand file I cut a‭ “‬V‭” ‬groove on it.

The spoke will be pulled into this groove and so locked at‭ ‬90°‭ ‬to the edge of the plate.

Next I took a length of studding and then drilled across it’s diameter close to the end,‭ ‬the drilling being just clearance for a spoke and an inch length cut off the studding.

Cross drilling studding

Component parts of tool

Component parts of tool

The spoke was then put through the hole,‭ ‬the cut studding put through the plate so that the chisel edge was against the spoke and a nut and washer fitted so that while the spoke was held against the plate it could still be pushed up and down through the bolt.

Assembled tool

Assembled tool

The plate was then put in place on the cylinder head with the spoke going down the plughole and the bottom edge of the plate was held firmly down onto the flat face above the plug hole.

The motor was then gently turned over and as the piston rose up the bore it pushed the spoke up through the bolt.

Tool in use

Tool in use

Since the bolt had been tightened enough to hold the spoke but not enough lock it in position,‭ ‬as the‭  ‬piston went across TDC and back down the bore on the other side,‭ ‬the spoke was left at the TDC position.

The device was removed from the head and a mark was put on the spoke at the chisel edge.

Reference was made to the bikes manual in the POC websites‭ “‬Library‭” ‬to find the correct advance distance for my motor,‭ ‬45°‭ ‬or‭ ‬19/32‭”‬,‭ ‬and another mark made on the spoke at that distance above the TDC mark.

‭(‬If you are making up one of these be careful to check what the correct figure is for your bike,‭ ‬official figure varies by some‭ ‬10°‭ ‬according to the year of the bike.‭)

This mark was aligned against the chisel edge on the plate and the nut tightened down to lock the spoke in place.

The motor was now brought to TDC on the compression stroke and then turned backwards to take the piston back down the bore to before the ignition point,‭ ‬the new tool put in place with the spoke end down through the plug hole and the motor gently turned forwards until the piston was felt to touch against the end of the spoke.

The motor was now at the correct spot to set the magneto timing at full advance so the Magneto could now be set in place and its timing set.

This is a bit more sophisticated than the old idea of using a‭ “‬pencil down the plug-hole‭” ‬but it’s easier to use when set up,‭ ‬it has a clearer index point AND there’s no chance of losing it down the plug-hole.

This new tool is small enough to keep in the tool-set carried on the bike itself, I’ve got an old tobacco‭ tin with the tappet spanners, push-rod tube spanner, mag spanners, a set of feelers, a packet of cigarette papers and now this tool kept in the toolbox, ‬after all,‭ ‬if you have it with you then you won’t need it but if you haven’t got it you will‭!

timing gears

Belt Up‭ (‬Part Two‭)

As the old lady was being laid up for the winter the chance was taken to drain and remove the fuel tank,‭ ‬this was rinsed out to remove any accumulated crap from across the season and the tank stored on a shelf out of the way.‭ ‬With the modern petrol brews having a limited shelf life it’s no longer a good idea to leave the tank full over the winter.

The carburetter was also drained so that the fuel in it could not evaporate and leave a varnish of additives choking the pilot jets.‭ (‬I was caught out by this last winter with the Velo and wound up having to put her carb in an ultrasonic cleaner to clear it out‭!‬.‭)

Next was to remove the magneto and dynamo and shelve these also for now.‭

This had cleared the decks so next step was to remove the timing cover.


timing gears

Open Timing Case

The joint was initially just cracked open so as to drain the oil and then the cover was removed,‭ ‬forgetting of course to catch the oil pumps pressure pad and spring as they dropped into the drained oil‭!‬.

Having fished them out of the drain pan they were wiped clean and put to one side and I found that unlike on the later motors,‭ ‬the magneto timing gear on my bike is not drilled and tapped to take a puller,‭ ‬it needs a claw type puller‭ ‬and, with the one I have, the legs foul against the timing case wall and prevent it getting a straight pull on the gear.

However on the older Panthers, such as this one, rather than the outer drive disc and the shaft being in one piece the shaft has a separate drive disc on one end and the timing gear on the other.

This meant that I could pull the drive disc from outside the timing case where there is easy access for the puller so it was just a case of unscrew the retaining nut and pull the disc.


Old Type Chainwheel, notice amount of wear!

Intention was to fit the puller and apply sufficient tension on it to hold things in place before the disc was heated with a gas torch.

Idea was to get the disc to expand a bit and then apply pressure before the heat was conducted to the shaft and expanded that as well, so it has to be done quickly.

Problem appeared in that the moment the puller screw was put in place and before any real pressure was put onto it, the threaded end of the shaft split in two! BUGGER!! Turns out that this is a not unknown problem on these motors, the shaft end being over hardened in manufacture.

Broken shaft

Shaft showing broken end

So I set to and turned up a pressure bushing on the lathe to press against the shaft end rather than the thread end.

Anyway, that worked, the disc was successfully removed and the shaft and gear could be removed from the timing case.

Next step was to replace the shaft with the converted one from the spare motor, only to find that it did not fit!. It seems that P&M increased the shaft diameter across the war years, TYPICAL!.

differing shafts

Two differing shaft diameters

Not only that but they altered the pressure angle of the gears so they are not interchangeable either.

This meant that I had to repair the damaged shaft, not so easy since it was through-hardened.

So the 4-jaw chuck was put onto the lathe and copper shoes put onto the jaws so as not to mark the shaft’s bearing surface.

shaft in chuk

Broken shaft in lathe-chuck

Using the 4-jaw meant that I could set the shaft to run dead true which was accordingly done and the broken end trued off.

The shaft itself was already centre drilled so it was just a case of opening the bore up to the appropriate tapping size.

Since the shaft was a hard steel I used a cobalt stub drill for this job, a cobalt drill will tackle far harder steels than an ordinary jobber drill will,  but their about three times the price!!

Then it was just a case of carefully tapping out the hole to take a bolt and the job was done.

I then had to transfer the alloy pulley over from the other shaft flange and I could then refit the shaft.

The shaft was then put in place in the timing case and the timing drive re-assembled.

The oil pump gear has a pressure pad and spring to hold it in place, this was stuck onto the centre of the gear with a blob of grease before fitting the outer cover.


Pressure pad in place before fitting the outer cover

Pressure pad in place before fitting the outer cover

On going to refit the magneto I ran into another two problems.

The first was that the new pulley masked the drive dogs on the bike side coupling.


New pulley masks coupling

To get round this the magneto had to be put onto its platform at the rear of the cylinder so that its coupling could be slid into mesh with the other one, BUT the studs that secure the magneto in place fouled the base plate of the magneto and prevented it from sitting down onto the platform.

Simple, you just remove the studs, put the magneto in place and refit the studs through the base plate! If only it were that easy!

P&M in their ineffable wisdom had chosen to use studs with an enlarged collar in the middle. This collar sits down into a counter-bore in the platform so it does not normally interfere with the magneto’s base, but this collar will not go through the mount slots on that base, equally the studs are threaded 1/4 Whitworth on one end and 1/4 Cycle on the other.

Short time remedy is to replace them with 1/4 Whitworth set bolts until I can source some double-enders but its not ideal as a Whitworth thread is more prone to backing off than the finer Cycle is.

While I had the magneto off I took the chance to replace the drive dog with the clubs easy adjust version and herein lay the other problem!.

The drive dogs on my bike, and apparently those on the later models using the “spider” type coupling, are 5/8 inch wide but the dogs on the clubs unit are only 1/2 inch wide, since my bike is among the first to use the enclosed valve motor, it looks like earlier models used a smaller size drive dogs.

Now it was just a case of fitting the magneto and re-timing the ignition.

Before fitting the magneto the endless drive belt for the dynamo had to be put in place, you can’t fit it afterwards, as I discovered!.

New drive

New dynamo drive

Then it was just a case of fitting the dynamo itself, tensioning the belt and refitting the drive cover. Once the cover is back in position the upgrade is invisible.

Dynamo installed

Dynamo installed

Cover intalled

With cover installed no change is visible


Back in the early days of biking (and of cars!) what you bought was the basic machine.

“Luxury” items such as lights were added after-market by the owner to suit their personal preferences and this meant that they came without any  electrics other than a magneto for the ignition.

For lights there was the choice between acetylene gas powered lights and electric ones, gas lights having the advantage of being a complete and separate system whose generator only needed to be kept filled with carbide and water but needed regular cleaning and refilling while an electric systems generator needed to be driven, a gas system was also cheaper than an electric.

A cycle type dynamo driven by one of the wheels was tried but found wanting so it went to being driven from the engine, but it also needed a “power storage unit” that is:- a battery.

Since it was an “add-on” this meant that the generator somehow had to be bolted on somewhere near the engine and then had to be driven from it somehow, even today on a car the alternator still is a “tagged on” part like this even after over 100 years of development!, but modern bikes have moved on from this primitive system.

However my Panther is a 1937 machine.

At this time, while bikes no longer came without lights as a standard fitting this was a recent innovation, as a “for instance” my 1929 Triumph was supplied without lights, they were still basicly a “bolt on” accessory.

The Panther has a 6 volt 36 watt dynamo bolted to the frame top tube, just in front of the seat tubes, and this is driven from the magneto’s drive coupling by a small duplex chain, this chain having a cast aluminium cover.


Position of dynamo

This drive is a known weak point on the bike, being prone to wear and possible breakage, and in the event of breakage it can cause damage to the engine housings and stripping of the timing gears in the event that the broken chain jams the magneto drive.

Dynamo drive

Dynamo drive chain

While it was not available back in 1937, there is a safer modern alternative available in the form of a toothed belt drive and I’ve decided to go this route since the dynamo chain on the old girl is showing signs of wear, as are the drive sprockets.

Investigations on the internet soon showed details of what was available in the form of toothed belt pulleys and drive belts as well as sources of supply of these.

On the Panther the drive sprocket has 30 teeth on it while that on the dynamo has 11, giving a 2.7 to 1 “gear up” in the drive, a maximum RPM for the dynamo in the order of around 7,000 RPM and of some 2,500 RPM at 30mph.

Drive Sprocket

Drive sprocket

I have fitted an electronic 12 volt converter onto my bike and since a downside of this is that it has a higher dynamo “cut in” speed than at the original 6 volts I’m going to take advantage of the conversion to uprate the dynamo speed a little and have decided on the use of a 14 tooth pulley on the dynamo and a 44 tooth as a driver.

This will give a dynamo speed of nearly 3000 RPM at 30mph and a maximum of around 8,000 RPM. From the use of these dynamos on other marques of bike I know this is well within the capabilities of the dynamo.

So I ordered up a pair of pulleys and a suitable toothed belt, which arrived two days later.

The small pulley turned out to be the neat belt width, there being a pair of cheek flanges to retain the belt in place and it was made from steel while the larger pulley was a wider, aluminium, unit and without the flanges so this will allow for a small measure of misalignment between the pulleys. Both pulleys were supplied with a “pilot bore” in their centres.

First thing was to sort out the drive pulley so as I have a spare drive coupling I decided to use that for the conversion.

This had two chain sprockets, one riveted either side of a flange to give a duplex sprocket. These rivets heads were accordingly ground off using a Dremel tool, the rivets driven out and the sprockets removed.

Once the coupling was offered up to the larger of the pulleys it was found that this flange was slightly smaller in diameter than the pulley and the rivet holes matched nicely with the side of the pulley and so could be used to mount the two together.

The pulley centre was then bored out on the lathe to match the register on the flange that had formerly located the outer sprocket. It was then counter-bored to give a good running clearance for the magneto coupling.

New Pulley

Coupling inside new pulley

This coupling is a variation on an Oldhams coupling which has the advantage that it can cope with the two sides being a little out of alignment in both plane and angle but it does needs some room to “work” to do this.

Once the pulley had been bored out it was fitted onto the location flange and the position of the rivet holes marked out on it.

The original rivets were 1/8 inch diameter so it was decided to use 3mm Allen cap head screws in their place so the pulley was now drilled and tapped to suit and then secured in place, the screws being secured in place with Loctite.

New Pulley

Magneto drive shaft with new pulley fitte

That left the dynamo pulley.

The dynamo has a tapered shaft for this to mount onto. This taper is a standard size across all makes of dynamo, it also is the same as that used on magnetos and is a taper of 1:10 on the diameter.

To machine this the top slide on the lathe needs to be set to the correct angle.

To do this I set up a test bar, known to be true, in the lathe and the top slide was adjusted to run true to this.

A 5 inch long ground tool-bit was then put in the tool-holder on the top slide and this was adjusted parallel to the test bar.

The top slide was then reset so one end of the tool-bit was against the bar while a ¼ inch drill was just nipped between its other end and the test bar.

This meant that the top slide was now set to cut a 1:20 taper on the radius, 1:10 on the diameter.

I then chucked up a length of half inch bar, cut a trial taper on it and tried an old chain sprocket on it.

Needless to say it was not quite right but given another couple of tries, making small adjustments to the slide setting each time, and I had a satisfactory result.

All that was left to do was machine the new pulley to suit and all was ready to fit to the bike.


To Be Continued

First Offering

It’s been quite a while since I’ve added anything to this blog, main reasons have been idleness on my part and having been out riding around on the Panther and her stablemates.

I’ve now begun another stage of the project and am prompted to make report.

It’s always been my intention to put a sidecar onto the Panther since doing so will allow me to take my dogs with me on the events that I take her on, as it is, when I go to an event  I have to either use the more modern BMW outfit or leave them behind which is not always convenient.

The sidecar I intend to fit is this Steib

Sidecar 2

As you can see it’s in nice condition

but when I offered it up to the Panther it was immediately obvious that they were not compatible.

Problem was that the Steib was on a chassis dedicated for use with a BMW, the major fittings were in fixed position with no adjustment possible and not only that but they presumed the presence of a front down-tube on the frame.

A problem here straight away because the Panther does not have such a tube!, it uses the engine as a stress bearing member in its place.

Anyway, what it meant was that with the rear fitting coupled in place the front one wanted to be through the timing chest on the engine and undoing the rear fitting and moving the chassis forward showed there was no-where for the front fitting to pick up to when it was moved far enough forwards to allow it to clear the timing chest.

So a change of plans, if the Steib chassis wouldn’t fit then I needed one that would fit, but that would also take the sidecar body.

I began a hunt for a suitable chassis and determined on a Watsonian VG21, a chassis of the same period as the Panther that would also take the Steib body.

Last time I wanted a chassis, everywhere I went I was tripping over VG21’s, but now I wanted one they had become like hens teeth!

I eventually tracked one down and work began.

First thing to do was to convert it to fit on the right hand side of the bike.

Reason for this is that both my other outfits (both BMW) have right-hand sidecars so having another one with a left-hand sidecar is asking for problems.

While doing this I found that the Silentbloc bush that the suspension arm runs on had been replaced with a solid steel bushing and the suspension locked up solid!

After removing this steel bush, and it did not want to be removed!!, I next had to source a replacement Silentbloc but I had a bit of luck here because I found someone who had the correct bush in stock.

So, conversion done and wheel fitted (I’ve not even looked at sorting out the sidecar brake yet!) it was time for the first attempt at offering up the chassis to the bike

At the moment I only have the rear main fitting sorted out so with this loosely fixed onto the sidecar it was connected to the bike and some wood blocks strategically placed.

A bit of shuffling it round and use of a tape measure and there it was in approximate position against the bike.


Sidecar chasssis

First offering up

I was now able to get some other measurements I needed then it was all again dis-assembled.

Main purpose was to check the feasibility of a right hand chair on the old lady, I was fairly sure it would be ok as there is a proper rear mount point for a sidecar at the offside rear , and to get the dimensions for having a swan-neck made up for the job.

This will be a simple piece of pipe-bending, a length of 1¼ inch OD heavy-wall steel tube with a 90 degree bend in the middle, simple to make if you have the facilities, but I don’t so it has to be farmed out.

Getting Tooled Up (3)

I now had the raw mouldings in hand but they were open backed.

To make the backs first the door apertures had to be trimmed to size and shape, a Dremmel tool with a cutting disk proved useful for this.

Next the rear sides of the boxes needed to be trimmed and sanded off level.To do this two sheets of 40 grit sandpaper were clamped onto a piece of MDF board and the boxes rubbed down on these.

Once I had the rear sides sanded down level and to the desired mark but before I moulded on the backs I needed a pair of doors for the boxes since it would be easier to sort out fitting the doors with the backs open.

The doors are a simple flat casting made on a flat board. They are reinforced with a plywood core.

If you remember, the door recesses in the end of the main mould were made by casting over two pieces of 3mm plywood cut to the shape of the door and fixed onto the end-plate of the mould.

These were taken off the end-plate and cleaned of the parting agent. They were then given a good sanding, about  1mm was then sanded off each side to reduce them in size a little and these became the cores for the doors.

A piece of MDF board was given several coats of wax and then a layer of glass cloth laid up on it and a core put in place on top of it.


Door core in place on first sheet

I had got some small cabinet hinges from the local DIY shop and to help secure these hinges to the doors each had one of their tongues enlarged by having a metal semi-circle soldered onto them.

The bottoms of the cores were shaped to accept the hinges, a pair were put in place and another layer of glass cloth laid up over them so there was now a sandwich of glass fibre/ plywood/ glass fibre.

Once the door mouldings had cured out they were trimmed to size and trial fitted in place so the toggle latches that secure them could be fitted, slots being cut to accomodate the hinge “tongues” at the bottom of the door apertures.
With the doors fitted I next had to make the backs.

The inside of the boxes were well cleaned down with acetone and they were then clamped down onto a flat working MDF surface that had been well waxed and then coated with a parting agent.

A bead of plasticene was put round the outside of the moulding/base plate joint to seal it.

The outside of the moulding wasnext covered over in case of accidental splashes of resin onto it and then, working through the door opening, a GRP back was laid up from inside each box.

The problem with these boxes however was that they looked what they are, a GRP moulding and the way round this was simply –  –  –  paint!.

So they were rubbed down and then painted so that they would match the finish on the rest of the bike and would look the part.

Once they were painted then the doors were finally fitted by sliding their hinge plates into the slots cut in the bottoms of the door apertures of the fibre-glass castings, being secured in place with that old faithful, Araldite.


Completed toolbox ready to fit on bike

Rather than mount the boxes with clips to the mudguard stays I made up back-plates from 16 gauge steel sheet. These were fitted onto the stays and the boxes bolted onto these back-plates using spreader plates on the inside of the boxes.

Once the boxes were bolted in place the bike was put on its wheels and a little water was poured inside them to find the low point and a small drainage hole drilled there so that the boxes would not fill up with rain water when out in the “inclement conditions” which so often characterise our riding season.

Finally here’s a lookat how the new boxes appear when fitted to the old lady.


Toolboxes fitted onto bike

I got all my filler and glassfibre materials from East Coast Fibreglass Supplies in South Shields <>.

Their shop is near enough that I just went over there and dealt with them direct but they do mail-order as well and I found them very helpful with advice. The hinges were small cabinet hinges I got from the local DIY shop and the latches came from, of all places, Screwfix!.

Getting Tooled Up (2)

Since the mould was to be made in sections next thing was to provide for this so a pair of plasticene “dams” were placed, one at either end of the top run of the plug, the surface cleaned off with acetone and then this section painted with a parting agent so the moulding would split off the base more readily.

Plug ready to use

Plug ready to use with the Plasticene “dams” visible, the blue tint is the parting agent.

First thing was a layer of gelcoat, this is the consistency of jelly paint and sets to give a smooth gloss surface against the plug or mould so about 75ml was poured into a mixing cup and 2% catalyst added.

To get the amounts right I put the kitchen scales inside a plastic bag to keep them clean and weighed the resin used.

The two components were thoroughly mixed together and the resulting gel brushed onto the prepared section of the plug.

Gelcoat takes around an hour to set so the job was put to one side in the warm and left to set.

While this was going on I prepared for adding the support layers. Since the plug was 100mm wide I cut 4 strips of glass mat, each 100ml wide and long enough to cover the section, I also cut two pieces of thin plywood 20mm wide by 100mm long to reinforce at the joints.

Once the gelcoat had set next thing was adding the first support layer. These layers are of glass mat reinforced resin so as I already had the mat prepared it was now a case of “Mix the resin”.

About 120mls of the laminate resin was poured into a mixing cup and weight checked, 2%of the catalyst was added  and the whole well mixed.

A generous layer of this resin was then brushed over the gelcoat and the first strip of the glass fibre mat laid onto this.

More resin was added to the topside of the mat and worked into the mat so that it was in close contact with the gelcoat, a roller made this job easier and the plywood reinforcing pieces were put in place at the two “dams”.

A second layer of glass fibre was then added over the first and “wetted out” with resin as before and the job put aside to allow the resin to set.

First layer on plug

Plug with first layers of mat added, the plywood end reinforcements can readily be seen

Once it had hardened then the whole thing was repeated to add another two layers of the glass reinforced resin and the job left overnight to setup.

The plasticene dams were then removed and the areas where they had been was cleaned off with white spirit to wash away the plasticene.

Another dam was put halfway along the bottom run of the plug so that the final mould would be in three pieces.

The ends of the first section were tidied up by removing surplus glass matt, the section wiped clean with acetone, coated with parting agent, laid up with glass fibre and left to harden.

Then the last plasticene dam was removed and the whole rigmarole repeated  for the last section, after which the whole moulding was allowed to dry and harden overnight.

Two locating holes were next drilled in each of the section ends, where the dams had been, and then the screws securing the side pieces removed and these were knocked off.

A plastic wedge was used to separate the pieces of the mould and these were cleaned up with acetone and polished.

Now the mould for the door face had to made up.

For this a piece of plywood was cut to fit neatly inside the mould and then another piece cut to the same shape but 12mm smaller all round. This smaller piece was centred onto the first and secured in place.

Door face mould

Mould for door face

A fillet of Plasticene was then put round the join between the pieces and to make it a concave fillet it was smoothed over with the ball end of an old pushrod. This gives a nicely rounded edge to the recess the door will fit into.
This assembly was screwed onto a piece of MDF board and the parts of the main mould then bolted together round the “door” piece.
A Plasticene fillet was now put round between the main mould and the door, again being finished with the pushrod end.

Mould ready to use

Mould ready to use

Procedure to make the toolbox moulding itself was much the same as for making the mould, wipe over with acetone, a coat of parting agent over the inside of the mould followed by a layer of gelcoat.
Once that had set the mould was then layered with glass mat and resin and left to cure.

Toolbox casting in mould

Toolbox casting in mould

The bolts holding the mould together were then removed and the mould peeled off the box moulding, the face piece was removed and I had the raw moulding of the toolbox.

Raw MoultingRaw casting just out of the mould (the white is the plasticine fillets)

To make the other box it was just a case of transfering the smaller part of the end piece from one side to the other of the main piece, re-assembling the mould the other way up and I then had a mirror-imaged pair of toolboxes just needing their door apertures cleaned up and the excess material round the moulding removed.