Reed valve

[Wesman07]

I woke up the other day with an epiphany. What if a four cycle engine had valve timing that suited the upper rpm range but did not suffer lower rpm power losses from intake revision? In theory, if there were no intake revision we could see a 30% increase in idling torque that would taper off to meet up with the peak torque. I wanted to bounce this idea off others before I dug into it much further. Physically adapting reed valves would be a challenge but it think its totally possible. Keeping them alive might be a little harder...?

After some research I found The 81 Honda xr 500 had reed valves so they could use a larger cam but maintain low end mannerisms. Being that this was only in production for a year or two, maybe its not worth the efforts? With the 300 motor needing such large cam profiles to get any decent HP out of them, it seems like it would be the perfect candidate.

 

[pmuller9]

It has been my own experience in the two stroke realm that adding a reed valve greatly improves low rpm torque but impedes the flow for high rpm horsepower as did the four stroke Honda xr500.
Secondly the 300 six make great power from 1200 rpm with the largest of the hydraulic cams.
As you go even larger in cam profiles the low end torque suffers more from valve overlap than intake closing timing.

 

[Wesman07]

You make a good point. The addition of reeds probably best suits low but very broad power bands. Its could also be a great loop hole for NA only racing classes.

I attached a screen shot of what I'm looking at. The four overlay graphs show stock, my current build, my current build with 100% anti revision and my build with a 230 duration cam and 100% anti revision. Note that the 230 degree cam and reeds are expected to produce 50 more ft/lbs at 1,200 rpm than production figures, 390 ish ft/lbs peak and just over 280Hp on 87 octane.

The rectangle intake ports in the head make a great location for a reed cage. The port could be opened up at the entrance in order to fit the largest reed cage possible. A quick internet search makes me think the 200 cfm would be not be a challenge but I haven't ran the math on anything yet. IF a thick header flange was used that covered the intake port, the reed cage could key into that flange. The intake manifold would bolt to the header flange and pinch the reeds in place. When its time to check the reeds and adjust valves, the upper and lower intake can be removed with out touching the headers.

 

[pmuller9]

The gains look good.
It would be a great experiment.

The one thing I forgot to mention was, when I installed the reed valve assembly on my 350cc two stroke engine there was a very significant gain in torque from an idle on up to midrange power but I was no longer able to use pump gas because of the increase in the DCR.
It cost me a piston before I got everything right.

 

[Wesman07]

I was thinking about that situation yesterday. Will reeds actually increase DCR on a two stroke? Won’t the cylinder still bleed pressure out of the exhaust port until it’s covered by the piston, at least until the exhaust pulse comes into tune?

Your first hand experience is the hard part I was talking about. For some reason the DCR isn’t showing a change in the general calculations. It is however showing up in max cylinder pressure with an increase from about 650 to 800 psi. I should have enough info to connect the dots.

Is there any advantage of moving the reeds away from the intake valve? Intake revision could be used to pressurize a long runner. I haven’t thought that one all the way through.

The new ktm 500exc have reeds. It looks like they are in between the throttle body and air filter. I wonder if that’s more for emissions?

 

[pmuller9]

My 350 two stroke engine was a Bultaco Alpina designed for low rpm woods riding.
The exhaust port roof was low in the cylinder for low end power and was only rated ta 22 hp.

The problem was the open piston port intake was not getting good cylinder fill at low rpm because of reversion.
I drilled four 1/2" holes in the side of the piston so intake flow occurred during the entire upward travel of the piston which reduced pumping losses and used the reed bank to create transfer compression.
Now the intake side matched the low exhaust port angle and cylinder fill at low rpm was impressive.
I could launch the front end over fallen trees from an idle.

If you move the reed valve away from the intake valve, the reversion will pressurize the added volume but the added volume will decrease the cylinder pressure at low rpm.
Until the intake valve closes the dynamic compression ratio is based on the cylinder volume plus the volume between the intake valve and the reed valve.

 

[Wesman07]

Smart. You were ahead of the times. What you did is still being done today. The only difference is that the pistons have an intake window cut in the skirt but its the same concept.

One of the two stroke motors I rebuilt this year had the piston installed backwards. The owner said his kid has never been faster since the rebuild lol.

 

[sixtseventwo4d]

One of the little things that made differences with flow using reeds was the angle of the plate they were installed on. And the material they were made from. Going from a steel type to a fiber was significant in that it helped with "bounce". Caging them helped with flutter. In all instance reeds wear a lot. I'm defining a lot from being used on 2 strokes. I remember a top reed company being called Boyesen. They even had two stage reeds.
Sure liked the way a 2 stroke would yank you when it hit the power band

 

[pmuller9]

If you move the reed valve away from the intake valve, the reversion will pressurize the added volume but the added volume will decrease the cylinder pressure at low rpm.
Until the intake valve closes the dynamic compression ratio is based on the cylinder volume plus the volume between the intake valve and the reed valve.

To get maximum benefit the reed valve needs to be as close to the intake valve as possible.

I'm thinking VVT would be a more reliable method of accomplishing the same result.

 

[Wesman07]

Boyesen is still making competitive reed blocks today and is 1 of the 2 quality aftermarket companies. They still use a design that keeps the feed laying flatter in regards to flow direction. Boyesen reeds probably flow the best.

The other company is moto tassinari. They take an opposite approach that uses more reeds at more of an up right angle. They are typically the most responsive.

Being that most modern day 250cc bikes are rate at about 50hp, six of them should be capable of 300hp. I did find some V3 reed cages for cheap enough that it might be a reasonable experiment. https://www.ebay.com/itm/224474612709?chn=ps&mkevt=1&mkcid=28

I had toyed with retrofitting VVT, but nothing really spoke to me for some reason. Maybe I wasn’t looking in the right places. I’m all ears if you have suggestions.

 

[pmuller9]

Forget that I mentioned VVT. It requires electronic and hydraulic control and is already being done in other engines.
The reed valve doesn't require a control system and will be different on a four stroke engine.
I like a new and different approach.

The reed valve will operate at a much lower frequency on a four stroke 300 six than on a two stroke engine and should last longer.
The lower frequency of operation will also make reed response time less critical.
In fact if the reed hangs open after 4000 rpm it would be a benefit.

The reed valve assembly needs to fit inside the intake port to minimize the volume between the intake valve and the reeds in order to be the most effective.
The simulations were based on zero reversion which would mean zero volume between the intake valve and the reeds.
Any volume added in that area will subtract from the results.

 

[Wesman07]

Ok so there might be something to this. I measured the stock intake ports at 105cc with a port dimension of 1.52" x 1.31". There may be a fork in the road here.

If we were to fit a reeds into the intake port, there would likely be a noticeable loss in port flow... Unless the intake port can be opened up enough to make up for the walls of the reed block. Adding reeds into the port would probably knock the port volume down by half. Do you think this is possible?

Option B is to put the reeds outside the head and use reeds large enough not to impede port flow. If I'm looking at this correctly 105cc is 6.4 ci. A cylinder volume of 50ci receding back into a port of 6.4ci would still be 87.2% effective. Now that number is on the high side being that I did not account for above and below the reed tips and a ported head.

 

[pmuller9]

Reversion begins somewhere after BDC depending on rpm and stops when the intake valve closes around 70 degrees ABDC so the actual piston displacement for reversion is less than 50ci.

 

[sixtseventwo4d]

If this helps with your reed theory in relation to needed cylinder volume, some shop air compressors used them. I've only seen this in use on piston type compressors. Those function as a 2 cycle
In relation the 2 stroke HP, It is derived from rpms and firing every time.

 

[pmuller9]

You can test the effectiveness of the reed valve on a single cylinder without having to run the engine.
Just get one reed valve assembly of each type you want to test.

After the engine's break-in run remove the intake manifold and take a compression reading using a compression gauge on one cylinder as a base line.
Then install the reed valve and take another compression test and see how much compression you gain.
You can do this with a reed valve at different distances from the intake valve to compare effectiveness.

 

[Wesman07]

That is a good idea. Actual cold cranking psi will probably be a more accurate number to work with than the calculations. I wonder if I could achieve good results with my 216 Bullet cam retarded 4 degrees... I do have an assembled OXX motor in the garage that I can use as a test mule, and two carb heads, one of them is in good condition.

I spent a few hours looking up reed valve sizes for various dirtbikes and snowmobiles. The reeds for the ktm 65 measures 1.5" but I don't know the other dimension yet. The reed valve designs I saw for air compressors would either be a significant flow restriction or lack durability do to their design. I have an email into Boysen asking for help with sizing and flow data if available. One of my concerns is over loading the reed pedals with a late intake valve closing or too much static compression.

 

[pmuller9]

The reversion on a 300 six occurs low on the compression stroke and the pressure on the reeds shouldn't be very high.
The reeds have to handle the full downward piston compression on a two stroke engine which is much higher.

You could start with the oxx engine as a feasabilty model and if it looks good then you can test it with the 216 bullet cam engine to get actual results.
The later the cam closes the intake valve on the seats the higher the gains with a reed valve.

 

[sixtseventwo4d]

The reversion on a 300 six occurs low on the compression stroke and the pressure on the reeds shouldn't be very high.
The reeds have to handle the full downward piston compression on a two stroke engine which is much higher.

The intake reversion pressure from the 300 would be placed directly on the reed if it is installed at the head. In 2-stroke application it has the volume of the crankcase to help absorb the pressure. It would seem the direct reversion pressure may be a bit more harsh.

 

[Wesman07]

The intake reversion pressure from the 300 would be placed directly on the reed if it is installed at the head. In 2-stroke application it has the volume of the crankcase to help absorb the pressure. It would seem the direct reversion pressure may be a bit more harsh.

That’s the way I was looking at it. Typical 2 cycle crankcase volumes are about 1.5- 1.7 times the engine displacement. The reason for those volumes are for complete cylinder fill plus the charge that enter the exhaust, it has nothing to do with pressure on the peddles. It’s just uncharted territory. It’s something worth asking someone at boyesen.

 

[pmuller9]

If a two stroke engine's crankcase has 1.5 more volume than piston displacement, the compression ratio is 3:1

If the intake valve in a 300 six closes 70 degrees ABDC the piston has only moved up 1" from the bottom.
If the static compression ratio is 9:1 the compression ratio at 1" piston travel is 1.3 to 1
The reeds will see a lot less pressure in the 300 six.

If you add the 300 six port volume between the intake valve and the reed valve then the compression ratio is even lower.

Trust the math. It is not uncharted territory.

Note: I edited my calculations.