Unibody Mk4 Spitfire 6

[BiTurbo228]

Thanks man  the research isn't finished!

One of the bits of information I've heard a number of times about these engines is that they detuned the 2600 by fitting the camshaft from the 2300 as on the test bed the 2600 was making 151bhp which was uncomfortably close to the 155bhp of the top-spec carbed V8.

This bit of info just floated around for a bit until I stumbled across the late, great Oldtuckunder's thread on measuring a cam in the hand, and how published cam specs don't seem to stack up to real life at all. It even came with a spreadsheet with graphs and everything!

So, I bought a dial gauge and some v-blocks, pulled out the NOS 2600 cam I bought and set to measuring.

It won't be 100% accurate as I did it round my GF's house with her mirror-flat marble counter top and I forgot a bolt to centre the dial gauge with. I did manage to get it pretty close by pritt-sticking the disc to the end of the cam though 

Over the course of many ad breaks I plotted the cam profile. I've also managed to find someone who has measured the valve lift of various M20B25 cams (not sure on their method, I'd have expected a little less of a parabolic arc due to the variation in rocker ratio). I've crudely estimated the valve lift on the Rover camshaft by taking away the clearance specs from the lift of the cam (although I think they close up a bit when hot). Either way, it's the shape more than the absolute specs that's interesting:

It's a needly little stick! No wonder it only makes 136bhp (but surprisingly good economy). If you can get shims to close up the gap a bit (which I'm fairly certain you can as both shim and bucket dimensions are shared between the 2600, the Stag V8/1850/Dolly Sprint, the O-Series, the Saab 99/900 B-Series, and the AJ6) I reckon there's a lot to be gained with a reprofiled cam.

Working out the exhaust cam specs is a little tricky without a built up head to play with. Like the Dolomite Sprint, these engines have a single cam lobe to operate both the inlet valve directly and the exhaust valve via a rocker. The exhaust cam specs will be the same as the inlet ones, but the rocker ratio will lower the lift and the placement of the rocker will set the lobe separation angle (there's something about timing peak exhaust valve lift with the point at which the piston is moving up the bore fastest, which is to do with LSA, but not got my head around that yet).

I've found a diagram of the 2600 rocker ratio in this excellent period study of camshaft friction comparing the 2600/2300 engine with the Ford Pinto (surprise surprise, the 2600 is better).

From this I've tried to crudely work out the LSA and averaged rocker ratio in MS Paint 

I'm almost certain I've got it wrong, but it seems to be around 88.7 degrees of LSA (most engines seem to be ~105-110deg). The more I think about it, the more wrong I've got it (I can't do it from the diagram effectively as it's midway through its valve lift). What I really need to do is produce a proper moving bodies computer model or actually measure it on a built up head...

But accuracy aside, the cam on a 2600 is mild to say the least. Plenty of room for improvement.

I've also got a TriumphTune Sprint 90 camshaft which I'm planning to measure which might be a little more relevant to those with OHC engines. Again, it won't be accurate enough to reproduce a profile from the information, but it'll give an indication of shape.

Oh, and the reason why all of this matters is inlet and exhaust runner length (I know more about inlet length so I'll witter on about that). Ideally, you want an inlet runner that's tuned to the pulses created by the opening and closing valve. When a valve closes, the air that's still rushing towards it piles up against the back of the valve, compresses and springs back up the runner (creating a pressure wave). When it reaches the end of the runner, for reasons I haven't quite worked out yet it reflects back down the runner towards the valve. If it hits a closed valve again it reflects back, and so on and so forth until it meets an open cylinder (I'm not sure what happens at that point, whether it bounces around inside the cylinder for a bit or not).

If you time this right, you can pile extra air into the cylinder, creating a mild supercharging effect. The fewer times this wave reflects, the stronger it is, which is why long curly intake manifolds (or long trumpets on carbs) tend to produce more torque. It's hard to get completely wrong as the pulses tend to align with valve timings at a number of different points in the rpm range, producing a VE chart that looks a bit like this for different inlet lengths:

Where I expect the trick lies is in packing as many peaks in as you can for a given packaging situation (so you want to know the lengths that align with max rpm), trying to align a peak with peak hp, and trying to align a peak with peak torque.

To calculate what lengths you need to target certain rpms, you need to know the cam duration. I'd done this previously using the published figures on camshafts (I'll add a spreadsheet with a calculator on for doing this once I've tidied it up a bit). Unfortunately, as Oldtuckunder demonstrated, published cam durations are frequently completely fictitious. So you need to measure things yourself!

There are a few things I haven't quite cracked the theory of yet, like why there seems to be a dip in VE at higher rpm for longer intake lengths on the chart above, and whether you want to be timing your pulses for when the valve opens (to help with exhaust scavenging) or just before it closes (to pile in a little bit more air into a cylinder that's already packed). But I feel like I'm getting there in my understanding

[Escadrille Ecosse]

Blimey, you are thorough! 

Keep going 

[Gt64fun]

On 10/2/2021 at 12:02 PM, Escadrille Ecosse said:

A plan so cunning you could pin a tail on it and call it a fox...

...weasel!!

[Escadrille Ecosse]

And ferret to you too

But yes you are right

[PeteStupps]

On 10/4/2021 at 11:54 AM, BiTurbo228 said:

When it reaches the end of the runner, for reasons I haven't quite worked out yet it reflects back down the runner towards the valve

Very interesting stuff! You're going deep, deep, deep into the detail 

The bit about reflections is a fundamental property of waves when they meet a change in dimension or resistance. I can't explain very well why it happens, but it's something to do with conservation of energy and momentum and the fact that waves can't change abruptly. Your pressure wave has a certain amount of kinetic energy travelling down the inlet tract; when it meets a step change in dimensions at the inlet mouth the energy & momentum in the wave must be conserved (neglecting friction losses) but the wave guide abruptly changes size. Waves can't do discontinuity, so the intake mouth is like a physical barrier. When it hits the mouth the wave splits into two - one going out and one going back in, with the total energy divided between them in some ratio which depends on the dimension change.

So you've got a reflected pressure wave, which will interact with other pressure waves in the inlet to give constructive or destructive interference at the valve. The longer the runner, the more wave fronts will be present in it for a given rpm. So I think it trails off at high rpm because there is so much destructive interference. If you could keep revving to infinity I'm sure you'd see more peaks in the curve somewhere, but not anywhere useful for real life. 

We did a practical about this at uni, although it was electromagnetic waves. We had copper tracks on printed circuit boards, where the dimensions of the copper changed at various points. A gizmo sends a pulse into the copper track and then records what it gets back (time-domain reflectometry, it was called). Basically we could calculate quite accurately what reflections you would get in the simplest case, but as soon as there were a few waves interacting with each other all our calculated predictions were miles out - any measurement errors get compounded when there are lots of waves.

Pete

[Escadrille Ecosse]

Keep going gents. This is interesting 

[BiTurbo228]

Fascinating! Doubly fascinating that it also works the same with electromagnetism. I never got too far into physics in school, but the more details you find out the more you realise that a lot of stuff that appears completely different is produced (or at least governed) by the same fundamental things.

Interesting on the destructive interference front too. That would explain why things like the continuously variable intake on the Mazda 787b is required, rather than just trying to cram the longest possible intake on there.

The question then becomes how do you match the length of the intake to the characteristics of the engine? 

I suppose you could assume that the graph I've posted is accurate (which is an assumption, but probably the best I've got so far!). That might mean that for a Triumph 2.5 engine with max rpm around 6000ish something around 500ish mm if you fancied a lot of power in the midrange, or high 300mm if you fancied a bit more top end at the expense of midrange. What I don't know is how cam timing might affect these peaks. I expect it'll just shift them around a bit for a given length (or shift the length around a bit for a given rpm).

I've found a couple of different methods for estimating pulse resonance, and the one that makes the most sense to my mind is this one: https://honda-tech.com/forums/honda-prelude-4/**calculating-intake-manifold-runner-length**-get-out-your-vtak-calculator-yo-1963201/

It's not quite the whole picture as the diameter of the intake tract has an effect, but by playing around with another calculator (that I've long since lost) that included a diameter metric it seems to have a minimal effect. Changing the diameter by a couple of mm only shifts the length by 10mm or so. We're not looking for pinpoint accuracy as we don't want to time a resonance peak exactly on peak power, we just want one roughly near there. Unless it has a strong effect on the magnitude of the waves I expect we can ignore it.

Using the 266 degree measured duration of the 308778 2.5 PI cam (kindly measured by @JohnD) that gives us these lengths:

Two different options for peak torque, and peak hp at 5500rpm. The speed of sound varies depending on air density, but again only affects the measurements slightly. All lengths are in mm, and are the total inlet tract from valve to bellmouth/opening.

It's unlikely that you'll be able to get a 2.4m intake in anywhere, and it's likely that would have deleterious effects at higher rpm. But they do match roughly to the graph:

If you want to time a resonance for 5500rpm, it looks like you'd want something 590ish mm for a strong midrange (or possible 470ish mm), and maybe 390ish mm if you want a smidge more top end at the expense of midrange. I wonder what would make the faster engine...

That's if you want to time the waves for when the valve opens. If it actually works by timing it for just before the valve closes then you need a much longer intake to hit a given wave. I think a possibly better way of thinking about it is as an rpm range at which a returning wave will meet an open valve.

So, for a 591mm intake this is what that looks like:

So the 3rd reflection is pretty much out of reach. The 4th will meet an open valve between 5500rpm and 6650. The 5th between 4400 and 5350. The 6th between  3650 and 4450. And so on and so forth.

For a 391mm intake this is what it looks like:

5th reflection is pretty much out of reach, but beyond that the numbers are roughly similar (if a little more bunched together).

Because the ranges are so broad but it's clear there are peaks and troughs from the graph I assume there is a right time to have the wave hit the valve opening, but what that is I have no idea! I can see how hitting the valve as it opens would help with exhaust scavenging, but I can also see how hitting it when it closes helps get more air into the cylinder.

I expect I'll just make something that fits the engine bay and as close to the numbers the formula spits out as possible. 591mm is difficult but maybe possible in the Spit engine bay, 394mm is eminently doable.

[PeteStupps]

Nice! 

Just to clarify, when you refer to the 4th wave is that a wave that has travelled up and down the runner 8 times? 4 complete round trips, like? 

And have you seen this before? https://www.emeraldm3d.com/articles/emr-adj-length-intake/ 

It is a nice bit of empirical research using adjustable length intake trumpets, and could be a useful validation exercise if you can do any calcs based on the info they give. Then see if your calculated results bear any resemblance to the measured torque curves. 

I can't help suspecting this is one of those things where it's difficult to match the calculations to the practical results, because of all ways the intake tract deviates from being a uniform straight pipe, and all the secondary influences which are hard to account for. But I am a pessimistic sod, so take with a pinch of salt

[JohnD]

Very interesting, BiTurbo!

Your figures confirm, in vastly more detail, a graph that appears in Stone's Internal Combustion Engines, relating inlet length to benefit from flow pulsation:

Stone referenced Campbell's The Sports Car, published  1978, where he describes thie above as a 'rough guide'.  But he also shows a chart of "the effect of inatke pipe length on volumetric effciency  on a D-type Jaguar":

Sorry its a bit crooked!     This is remarkably similar to your chart!

John

Edited October 8, 2021 by JohnD

[PeteStupps]

... Another thing that occurred to me on the subject, is that the energy in each wave will reduce significantly with each reflection (because a significant portion of the wave carries on without being reflected, as well as energy attenuated along the pipe). That ties in with the larger peaks and subsequent trough for the longer inlet; if you've only lost say 3 reflections worth of energy instead of 6, the constructive and destructive interference will be more pronounced. 

[Nick Jones]

I remember long and high powered discussion on this subject on the Eng-Tips forum, which I found when researching the design for my FI manifold. I wasn’t the OP but it was exactly the question about runner lengths and plenum volume that I would have asked.

It was eventually put in perspective by by the resident “tell it as it really is” guy who suggested looking at the sort of runner lengths being bandied about (long) and then taking a tape measure to the engine bay……

His observation was that the longest runners you can physically get in there will usually be the best you can do in the context of a road car, unless forced induction, when it doesn’t matter very much…

[Escadrille Ecosse]

30 minutes ago, Nick Jones said:

His observation was that the longest runners you can physically get in there will usually be the best you can do in the context of a road car, unless forced induction when it doesn’t matter very much…

I remember reading the Dave Walker piece on this (and others) some years ago when I was building the Spitfire.

I ended up with 1" trumpets on the Webers for the reasons identified by John

[BiTurbo228]

2 hours ago, PeteStupps said:

Nice! 

Just to clarify, when you refer to the 4th wave is that a wave that has travelled up and down the runner 8 times? 4 complete round trips, like? 

And have you seen this before? https://www.emeraldm3d.com/articles/emr-adj-length-intake/ 

It is a nice bit of empirical research using adjustable length intake trumpets, and could be a useful validation exercise if you can do any calcs based on the info they give. Then see if your calculated results bear any resemblance to the measured torque curves. 

I can't help suspecting this is one of those things where it's difficult to match the calculations to the practical results, because of all ways the intake tract deviates from being a uniform straight pipe, and all the secondary influences which are hard to account for. But I am a pessimistic sod, so take with a pinch of salt

Yep, 4th wave = 8 journeys up and down the runner. Each 'wave' would be one run from the valve to the inlet pening and back to the valve again.

I'd come across that article when I first started looking into this stuff but had since forgotten about it! Matching their findings with calculations is quite tricky without knowing the genuine cam duration specs, but I'll see whether anyone's measured them for any of those engines.

I expect you're right on the ability to match the paper calculations to the real world. The idea I had was to get the ballpark right and then make an inlet with a silicone section to vary the length a bit until it seemed to work right. That was the plan at least!

2 hours ago, JohnD said:

Very interesting, BiTurbo!

Your figures confirm, in vastly more detail, a graph that appears in Stone's Internal Combustion Engines, relating inlet length to benefit from flow pulsation:

Stone referenced Campbell's The Sports Car, published  1978, where he describes thie above as a 'rough guide'.  But he also shows a chart of "the effect of inatke pipe length on volumetric effciency  on a D-type Jaguar":

Sorry its a bit crooked!     This is remarkably similar to your chart!

John

Intriguing! That does look very familiar. Useful again for adjusting the ballpark, and I might just be able to find the actual duration figures for an XK cam (they didn't vary them terribly much). What I'd like to do is try to find out which waves they're hitting with each peak in VE, which might give me an indication of whether they're timing them for valve opening or closing (or both).

1 hour ago, PeteStupps said:

... Another thing that occurred to me on the subject, is that the energy in each wave will reduce significantly with each reflection (because a significant portion of the wave carries on without being reflected, as well as energy attenuated along the pipe). That ties in with the larger peaks and subsequent trough for the longer inlet; if you've only lost say 3 reflections worth of energy instead of 6, the constructive and destructive interference will be more pronounced. 

I wonder if there's a way of estimating how much energy is lost per wave/reflection. Preferably without needing to know the volumes of each part of the system, but I suppose that's only a case of measurement...

1 hour ago, Nick Jones said:

I remember long and high powered discussion on this subject on the Eng-Tips forum, which I found when researching the design for my FI manifold. I wasn’t the OP but it was exactly the question about runner lengths and plenum volume. 
 

It was eventually put in perspective by by the resident “tell it as it really is” guy who suggested looking at the sort of runner lengths being bandied about (long) and then taking a tape measure to the engine bay……

His observation was that the longest runners you can physically get in there will usually be the best you can do in the context of a road car, unless forced induction when it doesn’t matter very much…

And that's where it really lies of course! How much intake can you actually fit in the car.

I do think part of the issue is relying on straight inlet trumpets. If you look at a lot of OE inlets they curve back on themselves to get the added length. It's a lot trickier with a carb that needs to stay level, but injection would let you make something a lot fancier.

I measured the 2.5 PI inlet port as being 88mm long to the manifold face. The PI inlets are 132mm long to the end of the alloy casting. I did have a measurement for the gasket face of the engine to the inner wheelarch on a Spitfire/GT6 but seem to have lost it! If I find it I'll post it up.

I seem to remember working out that if you have a set of intake pipes that start from the PI inlets and spin around underneath them, meeting up in a plenum that runs underneath the PI manifolds (but above the exhaust) then you can just about get a 500ish mm inlet tract into the engine bay (although I never got round to testing the cardboard mockup with the throttle linkage in place).

1 hour ago, Escadrille Ecosse said:

I remember reading the Dave Walker piece on this (and others) some years ago when I was building the Spitfire.

I ended up with 1" trumpets on the Webers for the reasons identified by John

Yeah having carbs in the mix would make it tricky. You need a distance between the head and the carb to adjust the spacing, which would take up valuable room needed to get the bend radius in to duck underneath.

[BiTurbo228]

Oh, and another thought that's occurred to me after you mentioning longer intake tracts and deleterious effects at higher rpm Pete.

Why have I never heard of BMW people doing any of this?

People throw thousands upon thousands of pounds/dollars/whatever currency you fancy at BMW straight sixes of all varieties. From humble M20s up to fire-breathing S54s. Not once have I seen a properly designed 6-3-1 manifold. These eye-wateringly expensive, bundle-of-snakes, POA manifolds for an S54...are still a compromised 6-2-1 design.

How is it that us Triumph cheapos* have worked out that a 6-3-1 is much, much better than a 6-2-1 but the BMW fraternity haven't?

Perhaps it's something to do with the deleterious effects of pulse resonance at higher rpm (which I assume functions similarly in an exhaust as it does in an intake). BMW I6s tend to rev to at least 7000 for the more pedestrian go-fast options, and I've seen plenty of 8000-9000rpm screamers as well. If having great long intakes means they have a lot of punch in the lower 4000 rpm, but go a bit gutless in the other 4000rpm of the rev range perhaps it's not worth chasing. The design parameters are less 'let's use pulse resonance to make more power' than 'let's stop pulse resonance interfering with making power at 8300rpm'.

Although I suspect it's still something they just haven't cottoned on to properly. Or the people who have are keeping much quieter about it as there's actual proper money to be made from making a fast BMW.

*no offence! It's something I like about Triumph folk, we tend to be reluctant to build stuff simply by throwing money at it, which is something I value

Edited October 8, 2021 by BiTurbo228

[Nick Jones]

1 hour ago, BiTurbo228 said:

seem to remember working out that if you have a set of intake pipes that start from the PI inlets and spin around underneath them, meeting up in a plenum that runs underneath the PI manifolds (but above the exhaust) then you can just about get a 500ish mm inlet tract into the engine bay (although I never got round to testing the cardboard mockup with the throttle linkage in place).

ISTR someone trying this on PI with rubber hoses?  Don't remember the outcome, but I think it was complicated by the fact that it did have a notable effect on the engines VE, which messed up the MU calibration, muddying the waters.  Can't remember who it was now.....?

[RedRooster]

Whats wrong with K&N stub stacks?

Just use them and get on with your build.

RR

[JohnD]

I think this idea that in the impulse travels up and down the duct is mistaken.   We speak of 'tuning', and like organ pipes, the length and diameter of the tube is critical.    But all that travels the tube is a compression wave.  

This shows a 'fundamental' wave (top) with one half wavelength occupying the whole tube length. Next the first harmonic, exactly double that wavelength, and then three times that.

This lets you see how certain engine speeds, that drive a compression wave of a certain frequency, can resonate in a tube of a certain length.   And, of course, different frequencies at different lengths. 

The diameter isn't usually changed, much, but the length can allow this resonance to match the best part of the power curve and augment that.

John 

PS Gosh, we've got beyond the Common Room of Sideways University now!   This is Tutorial, or even Symposium territory!  J.

 

[DeTRacted]

I’m not sure about that John.  The diagrams you show are not of a travelling wave but a static or ‘standing wave’.  You will only get a standing wave if the excitation is continuous at exactly the right rate -  like blowing a flute.  I think an exhaust is much more like a tubular bell where the exhaust pulses are the stick.  There may be a resulting resonance in the tube but it would be more of a decaying sinusoid following each pulse.  As with a bell, the pulse repetition rate which does the excitation does not change the frequency of the resonance.

The exhaust pulse must travel - after all the gas does have to get out.  Another analogue is of sending an electrical pulse down a cable.   Provided the pulse rises quickly and is shorter than the transit time in the cable it will travel along largely unchanged until it meets a change in impedance (e.g. pipe diameter or joint in an exhaust). Then some of the energy continues but some reflects depending on the degree of impedance mis-match.   You can also see this with a skipping-rope held at both ends, where if  you tweak one end sharply you can watch the wave travel down the rope and then come back with reduced amplitude.  The faster the pulse rise-time and the shorter its duration the more pronounced any reflection will be so for an exhaust it should really only be a high-revs thing. 

The reflected energy will interact with following pulses in a complex manner depending on when they meet.  Surely tuning is the trick of getting the  pulse to arrive back at the valve with the right polarity and spacing  to give some rarefaction. ( I don’t think you can get a flute to suck at the player’s lips can you ?)

[JohnD]

DeTracted,

That there is a standing wave in the ducts is precisely my point!     And one driven by the exhaust valve opening and closing.   In an organ pipe, either a 'reed' , an automatic valve that does exactly the same, opening and closing to pulse the flow into the tube,  or  the shape of the lip above the air jet is such that the flow oscillates across it, as in a flute or recorder.    The pipe length selects the size and shape of the standing wave that forms inside it as the air column vibrates to produce the frequency and note that the organist wishes.   That standing wave is the base frequency for that tube, and the same note in the next octave is made by one twice as long, in other words, the length of the pipe tunes the organ - or the exhaust manifold!

The velocity of the exhaust gases may be estimated.      An engine is an air pump, and a four stroke engine will shift half of its capacity per two revolutions.    So at, say, 3000rpm a 2L engine will flow 3000 litres/minute, and 500L per exhaust port per minute.    If those are (again, say, as I can't be bothered to go out and measure some!)  2.5x3.25cms, then they have a cross-sectional area of 8.125cms^2 and the gas velocity through each of them at 3K will be 61538cms/minute, or 23 miles per hour.   

This ignores the effect of the heat of the exhaust gas, which again may be estimated.  At 800C, the density of air is about five times less than at room temperature, so the volume will be five times more and the velocity about 115mph.

This velocity is far from the speed of sound, but as anyone who has heard the Doppler Effect will realise, can influence the note, the frequency of sound travelling through it, as it affects stationary objects.       The observed frequency is calculated by Fe = (c/(c-V)Fo, where Fe and Fo are the effective and original frequencies, V is the speed of the gases and c is the speed of sound.   At 3K, the exhaust valve opens 25 times a second, so Fo is 25Hz; V is 115mph and c at 800C is about 1200mph.     The effect will be that the exhaust manifold will 'see' an impulse at 28Hz, thats a 12% difference.         We can't hear 28Hz, so here's (!) that a difference at  Middle A. It would mean a small (12%??) difference in the length of manifold primary to obtain a 'tuned' effect.

JOhn

 

 

 

 

John

 

 

 

Edited October 10, 2021 by JohnD

[DeTRacted]

I can't help thinking this would be better discussed  in the common-room but:

I understand your point John and am familiar with resonance and harmonics and with pulses in cables which is a good analogue as the physics is similar.   The reflection of fast pulses is much used in time-domain reflectometry and that goes for optics too because the same thing holds for fibres and light pulses.  Similarly, standing waves are set up in cables under the right conditions but these must be steady-state.

I can’t help feeling we are describing the same thing from different directions as it were.  Standing waves are an effect, not a cause and do not exist in isolation. They are the interference pattern from the interaction of incident and reflected travelling waves  (continuous sinewaves, not pulses) .  It follows that If there is no wave travelling downwards and no reflected wave travelling upwards, there cannot be a standing wave.

Considering each header pipe, I believe in practice any standing wave would be severely disrupted - often - by the short pressure pulses of gas, with a repetition rate and duration which will vary with engine speed  - so the conditions are far from steady-state.  There will be more pressure pulses occurring at different times and feeding from the other end of the pipe as the other cylinders exhaust so the result is likely to be very complex.   If it was one pipe per cylinder all the way down things would be simpler to comprehend.

[Escadrille Ecosse]

8 minutes ago, DeTRacted said:

 all the way down things would be simpler to comprehend.

It turtles all the way down as everyone knows

[DeTRacted]

Don't forget the elephants though.    (unseen university emoji)

[Escadrille Ecosse]

Never forget an elephant

[Nick Jones]

 Ah Terry, how will we survive without you.....? 

[BiTurbo228]

Thank Offler you lot are here  while I can follow along with what's being said it's beyond my expertise to give an informed opinion on it.

By all means keep discussing this here. It means I know where to find it when I come to doing something practical with it!

So, I wonder if the methods for calculating the desired length using the speed-of-sound-reflection method and the reverberation/pipe organ method come to something roughly comparable. I did find a number of other calculators nocking around the web, some very much rule-of-thumb guesstimates, and others of all sorts of complexity and methodology. I'll see if I can find one that looks like it takes your approach John. I seem to remember there being one about...

Edited October 11, 2021 by BiTurbo228