Rod’s Hoglet

[Roderick JenkinsParticipant @roderickjenkins93242]

Just completed the cylinders as the first stage of my Hoglet build. Of course, I immediately deviated from the drawings and made mine out of FC steel rather than C.I.

EN1A is wonderful stuff

They are not quite identical . I touched up the end of the parting tool after cutting the grooves on the first cylinder and because the tapered blade has a smidgen of top rake the grooves are a couple of thou narrower (and the fins correspondingly wider) on the second cylinder.

Rod

[Roderick JenkinsParticipant @roderickjenkins93242]

[JasonBModerator @jasonb]

Off to a good start Rod, are you going to just do the basic engine or will you include the kick starter?

[Roderick JenkinsParticipant @roderickjenkins93242]

Posted by JasonB on 22/07/2022 18:50:55:

… will you include the kick starter?

Not sure. I do like the idea of the engine being self contained rather than needing a drill to turn it over. I'll see.

Rod

[Roderick JenkinsParticipant @roderickjenkins93242]

The heads are made from c.i. and I coordinate drilled the hold down holes and left a 20 thou spigot on the head to locate in the cylinder. I've never been much inclined to make jigs and fixtures but, especially with 2 of everything, it seems necessary.

For drilling the cylinder fins for the hold down studs to pass through it all needs to be located and solidly held.

 

The long 3/16" centre drill worked very nicely

The heads need holes drilled and milled at various angles so I made this fixture today. The aluminium square is Loctited and screwed to a 16mm steel spigot that fits in the ER25 collet block. The square is centered on the spigot and has a locating recess for the head so, in theory, I can put all the holes in the right place using the DRO in the mill, tilting the collet block in the vice as necessary.

Rod

Edited By Roderick Jenkins on 30/07/2022 20:42:24

[GrahamSParticipant @grahams]

Good work Rod, looks spot on.

I am not too far behind you with my Hoglet build , I am finishing up the connecting rods then onto assembling the crankshaft then the bottom end is complete.

Like a lot of posts I see, I am not happy with the design method of a 'grab' fit of the crank webs to the shaft. I am planning using grub screws between the crank pin and webs to lock them, but cant decide on a method for the crank shafts to webs as after locking the pin there is no access. As a last resort I could just loctite them I guess, but not happy with this – Can I ask how did you do this assembly? Thanks

[Roderick JenkinsParticipant @roderickjenkins93242]

Hi Graham,

You are actually well ahead of me – I started withe the cylinders and am currently wrestling with the heads. I started with material I had in stock while waiting for the other stuff to be delivered.

I share your concerns but am a great believer in Loctite. My Wyvern crankshaft is fabricated only with the aid of Loctite 603 and has stood the test of time. Only yesterday I was disassembling a 1/2" PGMS bar in a 1/2" reamed hole fastened with Loctite and it took a lot of heat and a lot of hammering.

All the best,

Rod

[GrahamSParticipant @grahams]

This is where I am up to , with the bottom end complete except for the camshaft which I have been dreading.

The length of 3/8 bar poking through both crankwebs is an alignment aid I have built in , given I had little confidence in the pinch method of holding it all together while I pinned things together.

One of the crank pins was pinned first , followed by both crankshaft pieces. Then with the alignment bar in place, the other side of the crankpin was pinned. The pins are 3mm grub screws so removeable. When I am happy with it I will go back and loctite as well as pin. So far it all runs sweet so happy. Another advantage of this alignment bar is I can check everything is as it should be at any time , as in the picture the alignment bar passes through after assembly.

A view of the con rods in place before crankshaft assembly.

In this one you can see I extended the split in the crankweb to try and get more grip , to no avail

Please keep posting pictures of your build, thanks!

[Roderick JenkinsParticipant @roderickjenkins93242]

Nice work Graham.

I guess it might be useful to know what the Hoglet looks like. Here's a sketch taken from the Model Engine Builder article. It can bee seen why the inlets are set at half the V angle. The drawings are available in MEB issues 10 and 11 and available by PDF download for a modest fee.

I generally us an ER25 collet holder in my mill but in this case headroom is rather too tight so I used my Myford collets which fit the nose on the mill spindle. Once the DRO was zeroed in on the centre of the head I could use various stub drlls to make the required holes according to the drawing

The inlet and exhaust ports were drilled in the lying down mode and the relief for the push rods were milled with a ball ended mill. Trying to mill these reliefs with an end mill in the vertical mode did not go well, as evidenced by the gouges on the build plate – that was a scrapper

Inlet and exhaust ports are modified by counter-boring at the half V angle

Probably my fault but I could not find enough information on the drawing to work out where to start the angled counter-bore to intersect the inlet and exhaust holes so had to do some sketching to get a measurement to put the cutter in the right place

In this case, my 7/16" milling cutter happens to be on a 2MT shank which was a happy accident

The inlet has a 90 degree chamfer on the transition to the through hole so I had to make a D bit cutter out of silver steel

Quite pleased with the result

The spark plug hole is 10x1mm (M fine) drilled at 30 degrees. However it needs to be counter-bored 7/8" for tube spanner access. I usually use an Arrand boring head for this sort of thing but the bottom of the counter-bore needs to be flat for the spark plug to seat. So I dug out my Dore boring and facing head which I bought on ebay. I didn't have much hope for using the magnetic stand to hold the striker arm but in fact it worked very well

I returned the heads to the lathe to cut the fins. For the first time in the 10 plus years since I have had a VFD control on the Myford I had to use back gear. 50rpm was a nice comfortable speed with plenty of torque

That all got a bit tense towards the end but they turned out OK. I hope you find this mildly diverting.

Rod

[Howard LewisParticipant @howardlewis46836]

Splendid work Rod!

Should be quite an item when completed.

Just wish that I had your patience and skill!

Howard

[JasonBModerator @jasonb]

Looking good Rod, I think the 0.700" dimension is where the ctr line of the angled hole goes but it's a while since I did mine.

[Roderick JenkinsParticipant @roderickjenkins93242]

The crankcase is made from 4 bits of 1/4" ally plate. Front and back machined together, drilled and reamed for the crankshaft bushes

I made a bush from 1/2" PGMS to keep the plates aligned for further work

Milling the 21 degree angles for the cylinder seats (after cutting off most of the waste on the Lidl bandsaw)

Then hacking out the profile

and finishing with a 6mm slot drill and a 6mm ball end mill

The hole for the cylinder seat was step drilled in the pillar drill to 1" then transferred to the mill to use the boring head for final sizing

Then drilling the various holes

Tapping the holes on the side plates using the GHT pillar tool

I'm trying now to get a good join at the "ridge" so the side plates were fastened at the correct spacing to a base plate and the crankcase top plate angle machined in situ. With a bit of dead reckoning and a couple of swaps from side to side I was happy with the result

Cutting the "eaves" in situ as well

It's OK

More holes to make in the crankcase front for the cam housing and cam shaft. The spacing from the crankshaft to the camshaft will be slightly altered from the drawing because I will be using 0.8M gears instead of 32DP.

Rod

[Craig BrownParticipant @craigbrown60096]

Posted by Roderick Jenkins on 25/08/2022 23:21:44:

It's OK

Looks pretty seamless too me. If this is just OK then I look forward to a part you are very pleased with.

Thanks for the pictures and details, enjoying the build.

[GrahamSParticipant @grahams]

Looks excellent Rod, I will find the pictures for machining the heads very useful !

Looks like you are rapidly catching me up !

Having done everything else in the bottom end , I am going to try a set up for the cam shaft next , on scrap first as I am confident there will be a few practice runs.

[JasonBModerator @jasonb]

A couple of pics of machining the cam for mine if it's useful

[GrahamSParticipant @grahams]

Thanks for the pics Jason, that is a help. I think I just need to man up and make a start !

My main question is about the angular relationship between the cams, and which one is which !

Referring to the Camshaft Assembly drawing ( side on view ) and the end on view with the separation angles, how do you link the two together?

Assuming the cam furthest from the cam gear is the first one to be machined ( the orientation you used to machine with the long part of the camshaft in the collet ) , is the next one at 101 degrees further on or 113 or 214 ?

Mu guess is the first one is at roughly 8 o clock on the drawing , the large middle one at 12 o clock and the last one at about 4 o clock?

Hope that makes sense and thanks

[Roderick JenkinsParticipant @roderickjenkins93242]

My interpretation is that, since the drawing is third angle projection, the cam angle view is from the left side so the first cam is the 8 o'clock, the middle is 12 o'clock and the last is 4 o'clock. However, Jason has held his blank by the long end so the order of machining is reversed – 4 then 12 then 8.

I'm still having an internal debate as to whether to make the cams as described in MEB or use CamCalc. Since the base circle has to be machined incrementally anyway I'm not sure whether the MEB method saves much time since the nose radius has to be finished by hand. I have bought issue #7 of MEB so I have been tempted to try this method.

Rod

[Roderick JenkinsParticipant @roderickjenkins93242]

A digression on cams.

I've been slightly sceptical of the accuracy of the method that Randall Cox uses(and never really understood) to produce the cams (as shown in Jason's pictures) so I thought I would have a look at the process and compare it with the Manual Numerical Control method that is CamCalc. However, I must emphasise that the Cox method clearly works because there are many examples of running Hoglets.

I drew the Cox process (I loved my Spirograph when I was a kid). The large circles are the path of the cutter as it moves around the blank in 5 degree steps with the flank radius offset of 0.4" and a lift of 0.065"

 

A couple of points first: For a truly Harmonic cam the transitions between the base circle and the flank circle must share a radius, as must the transition between the flank circle and the nose circle. Although not given as a parameter in the Hoglet drawings, the CAD sketch shows that the action angle of the cam (the transition from base to flank) is 135 degrees. The nose circle is drawn as a tan, tan, tan circle to the the 2 flanks and the maximum lift line.

Using these parameters in CamCalc I get this:

The nose radius from CamCalc is a little larger than the Cox method shows (which is constructed by filing until it looks right). In CamCalc a sharper nose would be given by reducing the flank radius slightly.

The CamCalc method always seems a little laborious (not to say tedious) – take a cut, wind the handle,move the cutter height, take a cut, wind the handle, move the cutter height, take a cut… but the Cox method is no different really. I would use a 3 degree increment for this size of cam and Cox uses 5 degrees.

Conclusions: I like the Cox method, it has an elegant simplicity that does not require following a list of numbers. It is not quite as accurate as CamCalc in the nose area but for the sort of engines we make this is immaterial. For this particular engine there is a down side and that is that the cams must be machined in the vertical mode with the ensuing problems of rigidity with an interrupted flycut. With CamCalc the cam blank would be horizontal and the machining is a light cut with an end mill for each increment of rotation and no risk of interference from the boring bar on the subsequent cams.

I shall be using the Cox method for my Hoglet 'cause it's good enough and a different challenge and I like that.

Rod

 

Edited By Roderick Jenkins on 01/09/2022 11:49:46

[JasonBModerator @jasonb]

The Cox method also needs less blending of the facets than the CC as the fact they are cut to a large radius leaves a smoother finish.

Rod are you not tempted to do it on the CNC these days, I think the cut time for the last cam I did was about 70seconds which is far less than CamCalc . No need for final hand finishing and you can blend whatever radii you like for the nose.

You would probably have to use a tee slot or woodruff cutter to do the Hoglet one as a single part. Could even cut it in three pieces and loctite them onto the silversteel shaft, should be enough area for loctite if you did the outer cams integral with the spacers and then just the central longer one to length then they could be done with say a 6mm milling cutter.

[Roderick JenkinsParticipant @roderickjenkins93242]

Posted by JasonB on 01/09/2022 12:05:39:

The Cox method also needs less blending of the facets than the CC as the fact they are cut to a large radius leaves a smoother finish.

Rod are you not tempted to do it on the CNC these days…

Yes. Sadly, I dont think my Sherline mill with its maximum collet size of 1/4" is rigid enough to make the cam shaft in vertical mode with a woodruff cutter hanging 1 1/2" out of the chuck. If only Fusion let us have a 4th axis

I had thought about individual cams and cutting the gap between them so that they interlock at the correct orientation. But, the Cox method is new to me so it's always nice to try something novel.

Cheers

Rod

PS: I think the Cox method produces a smoother facet because the cutter has a smaller radius than CamCalc which is a straight line cut and therefore of infinite radius 

Edited By Roderick Jenkins on 01/09/2022 13:07:59

[Howard LewisParticipant @howardlewis46836]

Whichever method is used the finished product looks to a beautiful piece of machining..

Just wish that I had your knowledge and skills.

Howard

[Roderick JenkinsParticipant @roderickjenkins93242]

One thing I should mention is that I have enjoyed Andrew Whales' Hoglet videos on YouTube – very useful for envisioning the bits even if we choose different methods according to the kit we have available.

The cam housing is a complicated bit of carving

Need to bring out the Dore facer again to cut the space for the cam gear

The top of the housing is at 25 degrees and the cam follower holes at 23 and 29 degrees. The housing sits on a jig made from 2 bits of steel out of the oddments drawer Loctited together. I have marked every ten holes (1 degree) on the dividing plate with a Sharpie to help keep track of the rotation

Drilled 6mm and reamed 1/4". Fortuitously, the centre on my dividing head tailstock is 1/2", the same diamter as the relief for the cams to rotate in so I was able to provide a bit of support

The housing cover was squared up

Held by the cover to hack out some more swarf

Then trimmed

Forgot to take pictures of drilling and reaming the cam shaft hole and rounding the profile – kept getting distracted by the sounds of Merlin engines flying round the house

When I made the Crankcase front plate I drilled the smaller hole according to the drawing, forgetting that I needed to enlarge the dimension by 9 thou to cope with the slightly larger diameter gear caused by using a 0.8M rather than 32DP. So, squared it up

and re-centred

then drilled the holes and reamed in the correct (I hope) places

 

Just one more cosmetic oversight to trim

Phew. The cam shaft and crankshaft holes are oversized for bearings to be pressed in but they seem to line up, which is nice.

Thanks for looking in.

Rod

Edited By Roderick Jenkins on 04/09/2022 16:47:52

[Roderick JenkinsParticipant @roderickjenkins93242]

I think this is my favourite bit, making the gears

2 blanks required for 40T and 20T gears

Something weird happened which I have not seen before. As I parted off the blank with the rear tool it bounced on the cross slide and jammed between the tool and the holder. Bummer-that's for the scrap pile

Both blanks were faced to thickness. This is a trick I got from Editor Neil's "3D Printing for Model Engineers". It really works, the faces are parallel to 1/2 thou

I have learned through bitter experience that you cannot trust the profile of these Brown and Sharpe style cutters to be central to the thickness. I have not come up with a better method than eyeballing

Cutter running at 250rpm and a full depth cut. Each tooth is 1 1/2 turns for the 40 and 3 turns of the dividing head for the 20

Very satisfying

Rod

[Roderick JenkinsParticipant @roderickjenkins93242]

Cams. This is my setup.

For the top cam the table is at it's very lowest extent. A respectable cam set was produced though it was hard work. I try and avoid feeding a cut by raising the table ( I don't have a quill). 2 hours of moving the table up and down was painful on the wrist. Sadly, I got the orientation of the single cams reversed. I found it difficult to get my head around where the peak of the cam is using this method. The interrupted fly cut, together with the flexibility of the whole setup is not very restful either. I prefer CamCalc for the next try- much gentler.

3 degrees (half a turn) between each cut

The cams are made fro EN3B and case hardened. I am fortunate to a have a small stock of genuine Kasenit. The ends were covered in Tippex Rapid then the blank heated and dipped in the Kasenit before heating to bright red and quenching in water

The dappled grey is a good sign

A file skidded off nicely. The duff one (upper) and the good one polished with some wet and dry.

Rod

[Richard MillingtonParticipant @richardmillington63972]

A very interesting build.

I have learned through bitter experience that you cannot trust the profile of these Brown and Sharpe style cutters to be central to the thickness. I have not come up with a better method than eyeballing

Would the old ruler between the pointer and part work?

[Author]

Posts