[DailyDriver]

As for tension for the Gates blue belt from JAE, I have used them on both a 2001 V8 and a 1990 SE. I use the frequency method for tension on the belts using a laptop, a microphone and an app used by the Ducati community to tension their belts. To determine a frequency to use on the blue belts I made a fixture to compare the new blue belt to a new original black belt and also a used blue belt to a used black belt.

The fixture consisted of two V8 camshaft pulleys on spindles mounted the same separation distance as on the V8 engine. I would put a belt over the pulleys, hang weights at the bottom of the loop and then take a frequency measurement between the pulleys.

I went in 5 lb increments from 25 to 60 lbs. Frequency increased in a linear slope from about 90-94 hz at 25 lbs and ending at 140-145 hz at 60 lbs. The results were very close on all belts tested... new blue, used blue, new black and old black.

I found the same weights gave remarkably close frequencies for all the belts.

I now use the specifications as specified by Lotus for frequency tension on both the V8 and the SE.

 

[jtrealty]

While I admire your effort to try to build a test stand, I don't think it replicates properly all of the conditions. On the actual motor the cams exert a torque against the belt. Part of the necessity of accurately setting the position of the crank and cams prior to trying to measure the tension of the belt.
David Teitelbaum

 

[DailyDriver]

While I admire your effort to try to build a test stand, I don't think it replicates properly all of the conditions. On the actual motor the cams exert a torque against the belt. Part of the necessity of accurately setting the position of the crank and cams prior to trying to measure the tension of the belt.
David Teitelbaum

I have no clue what "torque" has to do with anything. What is measured by the frequency is the tension on the belt. Any of what you call "torque" on the cams only results in tension upon the belt. There is absolutely no difference between the weight exerted by the load and any "torque" applied to the pulley. Hence: torque on the pulley directly correlates to tension on the belt.

 

[Esprit2]

Well, I'm in David's court on this one. The whole point of the timing belt is to transmit strand tension force to overcome a TORQUE-load of drag in the camshafts that is caused by the valve train. Draping a belt over two pulleys that aren't driving anything, applying a tension load, and measuring a resulting frequency value doesn't confirm that enough force could be tranmitted via the tooth mesh to reliably drive the cams without jumping timing or stripping teeth... both of which are faillure modes common to too little tension. Too much tension is bad because it can break the belt. Too little tension is bad because the belt can jump timing on the pulley, or the belt's teeth can be damaged or stripped off of the belt. Any of which will bend valves in an interference engine. So the 'correct' tension is important !!

Once you know what that tension is, then it's nice to also know a correlation between pounds of strand tension (Burroughs) and frequency response (Hz). Correlation has value in that it gives owners more 'instrument' options. However, it doesn't establish how much tension is necessary to produce the drive capacity (tooth mesh strength) required to overcome the torque load of the cams & valve train in the first place.

And pounds on a Burroughs gauge is not the same as pounds on a Krikit KR1, a Trident KT20207... or on any other mechanical gauge. None of them directly measure strand tension in pounds. They all measure some other attribute, and then based upon laboratory testing, correlate that to the correct tension. So each different type of mechanical 'tension' gauge will all correlate to frequency differently.

Nice try... correlation has value. But determining the correct tension value first is critical... then correlate to something else.

Regards,
Tim Engel

 

[DailyDriver]

So you are telling me that when a belt is placed over two free-wheeling pulleys and weight is applied to the bottom of the belt loop to apply tension between the pulleys that is different than if one pulley was fixed and you applied a torque to the other one to achieve the same tension. I submit to you that is false. The only difference is one measurement would measure the weight applied and the other would measure the torque applied to the pulley. One is much easier. Physics doesn't care where the force for the tension comes from between two free-wheeling pulleys.

The purpose is to compare the four different belts to see how they compare in measurement against one another under the same tension and the same distance between pulleys. Lotus decided that frequency measurement is the method to determine the tension of the belt. I measured the frequency under conditions than are reproduced the same for each belt under the same conditions. If Lotus says a given belt is at correct tension by measuring the frequency between the pulleys then my frequency measurements apply to the same pulleys mounted the same distance apart.

With the difference in frequency measurement being negligible under identical situations between the blue and black belts, both new and used, between about 90hz and 140 hz it provides a measured basis for me to stick to the specifications in the manual for adjusting the blue belt. I have never heard anyone state any basis for what tension to set the blue belt that was based upon anything but a guess. This experiment shows that under identical conditions both belts appear to give the same tension based upon the Lotus approved frequency method.

So, what experimental method would be suggested for the sole purpose of comparing the tension of blue versus black belts. What are the recommendations for tension on the blue belts versus the black? Based upon what measurement?

I submit the Lotus specifications apply to the blue and black belts the same.

 

[Esprit2]

You submit wrong.

 

[jtrealty]

Expanding on "tension of the belt" that value is determined by a lot of things. When you set the tension the engine is cold. When the engine gets hot, it, of course expands, further increasing the tension on the belt. Then there is the belt itself. Some belts stretch, some don't and some more than others. New vs old, blue vs black, etc. The Lotus engine is a very expensive thing to experiment with, how courageous are you? Our best knowledge now is experience. There is a lot of history on how to tension different kinds of belts and the procedures to do it. I don't think your rig accurately duplicates the true operating conditions.
David Teitelbaum

 

[Esprit2]

So you are telling me that when a belt is placed over two
free-wheeling pulleys and weight is applied to the bottom
of the belt loop to apply tension between the pulleys that is
different than if one pulley was fixed and you applied a torque
to the other one to achieve the same tension.
I submit to you that is false.

You're right... what you just wrote explaining your understanding of what I wrote is False. Go back and read what I wrote again.

Are you aware that according to Lotus specs, the black HTD belt runs at a lower tension than the black trapezoidal belt? "No, that's not what the manual says!" Yeah, sure.

The trapezoidal tension went through an evolution before they got it right. In the end, it was 95 on a Burroughs BT-33-86J gauge that was calibrated specifically to the Lotus trapezoidal belt.

When the HTD belt came along, it's deeper tooth engagement required less tension. But since the Burroughs gauge is thickness sensitive, and the the HTD is thicker, the same 95 spec, on the same trapezoidal calibrated Burroughs, produced less tension in the HTD belt. Something closer to 90...or high 80s.

When Lotus introduced the frequency spec for the HTD belt, it was 110-110 Hz at 30 BTDC. If you set the HTD to 95 at TDC, which is really <90 at TDC, then rotate the crank around to 30 BTDC where the tension that varies with crank position is something else... what the heck are you really comparing the 110-110 Hz frequency to? Do you know? Or are you just looking at the dead weight on your test fixture, and saying, "the tension is constant"? Lotus didn't feel the need to explain all of that to owners... "Just follow the instructions".

There never was a Lotus frequency spec for the black trapezoidal belt, and it needs more strand tension than the HTD belt to function properly. Tension does not apply equally to them. So if you're using the same pounds tension for both trapezoidal and HTD belts to compare with frequency readings.... phbbbbt.

Gravity and mass are constant, and will maintain a constant tension in a belt on your test fixture. On the engine, things aren't that simple.

If the correct tension is known for each belt tooth form and construction (trapezoidal, HTD, HCR, HSN & HNBR), then your fixture could be used to correlate a trapezoidal tension in pounds at TDC, to a frequency equivalent at 30 BTDC. But simply hanging a dead weight on the bottom of a belt loop around two free pulleys isn't going to tell you what the correct tension is for that belt in order to survive running conditions on that engine.

 

[DailyDriver]

When installing a black belt on either a V8 or a 1990 SE and follow the procedures specified by Lotus you set the belt tension by frequency. No other value is provided by Lotus. The frequency is a direct indicator of the tension upon the belt. Lotus specifies one place for measuring the frequency and that is between the intake and exhaust pulleys, only. The correct frequency adjustment by adjusting the tension assures the belts wont skip, fly off or whatever. Hence, the belt is at the proper tension by Lotus definition.

Now when installing a blue belt on either a V8 or a 1990 SE the only specified procedure for the V8 is the frequency method and the updated Lotus procedure for the SE is also the frequency method. Lotus does not supply a frequency specification for the blue belt.

When installing either a black belt or a blue belt in either vehicle the identical setup procedures are used on each. Every pulley, every camshaft, every piston and everything else is in the exact same position for the frequency measurement. Any forces imparted on the belt by anything in the engine are exactly the same because the setup is the same, those all become constants. The only variable is a black belt versus a blue belt.

If a black belt is placed over two free-wheeling pulleys with the same separation as those in the car and a force (weight in this case) is applied to the bottom loop of the belt there will be a tension applied to the belt between the free-wheeling counter-rotating pulleys. Because, as Lotus demonstrates by their procedures that proper tension equals proper frequency then the inverse is true... proper frequency equals proper tension. If a force is applied in the bottom loop of the belt such that the frequency measured between the pulleys is the same as the frequency specified by Lotus then that belt is at the same tension between those two pulleys.

By placing a blue belt on the same fixture from the black belt with the same force applied to the bottom of the loop everything remains a constant and the only variable is the belt. By measuring the frequency the same as with the black belt you will be measuring the tension upon the blue belt, again the only variable is the belt.

Since Lotus only supplies the specification for frequency the question is does the blue belt obtain the same amount of tension as the black belt by measuring the same frequency? Does 120 hz frequency measurement produce the same tension on both belts?

When an identical force (eg. 40 lbs of weight) is attached to the bottom of the loop in the belt for each belt then the force will apply the same tension for each belt. There are no other variables, only a different belt. If both belts show the same frequency with the same force then the Lotus specifications for belt tension will be the same. If they show a different frequency for the same force applied then Lotus specification of frequency would not be the same from the black belt to the blue belt. Some characteristic of the belt itself would have to be explored.

We are not dealing with trapezoidal belts, no Burroughs Guages, simply what frequency to use when putting a blue belt on a V8 or SE when the only frequency specifications from Lotus are for the proper respective black belts.

Again I ask, what frequency is your correct frequency for the blue belt on the V8 and 1990 SE and what is it based upon?

 

[snowrx]

"We are not dealing with trapezoidal belts, no Burroughs Guages, simply what frequency to use when putting a blue belt on a V8 or SE when the only frequency specifications from Lotus are for the proper respective black belts"

Yes we are. This is a topic specifically mentioning the availability of another belt option for the older trapezoidal belt cars, and this inevitable discussion is exactly why I avoided the tension topic in the original post, as there are a number of posts on belt tension already. I agree tension is tension, and would love someone to give me a frequency for the T104RB on an early car, but we're off topic.

 

[Esprit2]

When installing a black belt on either a V8 or a 1990 SE and follow the procedures specified
by Lotus you set the belt tension by frequency. No other value is provided by Lotus.

Correct for the V8... there is only one Lotus tension spec, and it uses the frequency method.

However, you have repeatedly mentioned your 1990 SE, and for the 4-cylinder's entire production life, the Lotus manuals have specified the T-belt tension in pounds measured with a Burroughs gauge. The frequency method came along after 4-cylinder production ended and after Lotus embraced the frequency method for the V8.

For the black belts on the 4-cylinder, the Lotus tension spec was 95 lbs, measured midway between the intake and auxiliary pulleys, with the crank at TDC. So there is a pounds (Burroughs) tension spec for your SE that doesn't involve frequency, but I guess you have a choice to ignore it.

The BLUE belt is a completely grassroots application for both the V8 and the 9XX 4-cylinder engines. There is no factory support for the Blue belts from either Lotus or Gates. Lotus simply is not going to invest more development engineering time & money into an engine that went out of production many years ago... no matter how good the new part is. But having said that, when JAE first introduced the Blue HTD belts for both the V8 and 4-cylinder engines, Lotus Cars plc purchased inventory from JAE. They knew a good thing when they saw it, but they weren't going to invest time & money into past engines. There is no Lotus or Gates tension spec for the Blue belts.

The only variable is a black belt versus a blue belt.

True for your test rig. But that doesn't mean there shouldn't be more factors considered.

By placing a blue belt on the same fixture from the black belt with the same force applied
to the bottom of the loop everything remains a constant and the only variable is the belt.

You are presuming that all the belts are created equal, and they are NOT. Each belt tooth shape, construction & materials has it's own tension needs, and lumping them together under one tension is a mistake.

First, the belt tension must be correct for a running engine at normal operating temperature... ie, hot! That's a given and not debatable. But the Lotus tension specs are given for a cold engine that is off, since that's when you would normally work on them. As the engine warms up, it expands, stretches the belt, and the tension goes up. But each belt construction/ materials has different Tensile Strength, Coefficient of Elasticity, and Coefficient of Thermal Expansion... plus other factors. You're simply presuming one tension fits all (40 lbs on your test fixture). It doesn't.

Second, each belt type has it's own needs... forgetting what the engine needs. The Blue HNBR HTD belt is a quantum leap less stretchie, and requires a lower cold tension in order to produce the correct hot running tension. If you set the Blue HTD belt to the same cold tension used for the black HTD belts, the resulting hot tension will be WAY TOO HIGH. The Blue belts just do NOT stretch as much as the black belts.

And you choose to ignore all of the above.

When an identical force (eg. 40 lbs of weight) is attached to the bottom of the loop
in the belt for each belt then the force will apply the same tension for each belt.

So where did you get 40 lbs at the bottom of the loop as being correct for any belt? "IF" you cut the belt, anchored one end, looped it up over the pulleys, and hung a 40 lb weight from the other end, then the strand tension in the belt would be about 40 lbs (ignoring pulley bearing drag). However, by suspending a 40 lb weight from the bottom of a continuous loop, the strand tension is only 20 lbs.

For the 4-cylinder 9XX engines (your 1990 SE), Lotus specifies 95 pounds (Vs your 20 lbs) measured via a Burroughs gauge. Due to the Burroughs' sensitivity to thickness, the result is an accurate 95 lbs in the trapezoidal black belt, and a lower tension at about 90 (maybe high 80s) in the black HTD belt. So by what logic are you discounting Lotus' specs and presuming that a 40 lb weight in the loop (20 lb strand tension), is a correct tension? It's not.

There are no other variables, only a different belt.

Changing belt types automatically introduces a plethora of variables, which you conveniently choose to ignore.

Since Lotus only supplies the specification for frequency...

For decades, Lotus has specified the black belt tension for the 9XX 4-cylinder (your 1990 SE) in pounds-Burroughs... it's in the "Service Notes" manual.

the question is does the blue belt obtain the same amount of
tension as the black belt by measuring the same frequency?

The blue belt should not be presumed to use the same frequency as the black belt. Said another way, the same frequency in different belt types will not produce the same strand tension in them.

The blue HTD belt is thicker, heavier and stiffer than the black HTD belt, and those are all factors that affect frequency response. On a guitar, if you set both the skinniest string, and the fattest string to the same strand tension, then pluck them, they will produce very different notes. Going the other way, IF you want the same strand tension in both of those guitar strings, you can't simply set both of them to produce the same note, since the resulting tensions will be very different.

We are not dealing with trapezoidal belts, no Burroughs Guages,

Who's "we"? This thread started in 2016, with no mention of V8s through Snowrx's post two days ago. You didn't jump in until yesterday.

simply what frequency to use when putting a blue belt on a V8 or SE when the
only frequency specifications from Lotus are for the proper respective black belts.

Again I ask, what frequency is your correct frequency for the
blue belt on the V8 and 1990 SE and what is it based upon?

I've previously stated what I have found to be a serviceable tension in Lbs-Burroughs tension for a Blue HTD belt in a 910 4-cylinder (your SE). If you're not willing to purchase or borrow a Burroughs gauge to use that tension spec, then don't expect me to buy a frequency analyzer to convert the pounds to Hertz for you.

Gates Racing Blue HTD belt tension = Burroughs 83-85 = Krikit KR1 34-35 on the 'pounds' scale

I do not know a correct Blue belt tension for a 918 V8, someone else will have to post a recommendation for you. I hope you'll listen to them.

I'm done. Good luck.
Tim

 

[DailyDriver]

To SNOWRX. You are correct. This thread has gone a bit off topic from the original dealing with determining the frequency for your specific belt to a discussion of the validity of a test procedure that may be able to answer your original question.

I have heard for years that nobody can tell what frequency is appropriate for the correct tension of the blue belt. All I have heard is "I don't know" to a guess. Nobody seems to have even tried to measure and compare the black to blue belts in a static load and static tension condition as is found when testing the belt upon installation in the car.

I made a fixture using two camshaft pullys from a V8 spaced the same distance apart as they are in the head. The fixture allows a belt to be placed under the same static tension as specified in the manual. Once the belt is at the same tension as on the engine, i.e. providing the frequency specified by Lotus, then you can measure it any way you wish. You can measure the belt with a Kriket, a Burroughs, frequency or twist. Each will give their own reading in their own way for the same static tension.

If I were to do this test for the 4 cylinder I would first set the black belt on the engine to the correct tension using Lotus specification for that engine. Then I would set the engine to the specified parameters for the frequency method as specified by Lotus, 30 deg BTDC. Measure the belt using Kriket and frequency (Kriket because I believe you referenced it previously as your method) at 30 deg. BTDC as specified for the later models just before removing the belt. Both measurements at the exact settings specified by Lotus for the frequency method. Remove the belt and place it on the fixture. Apply enough load to match the Kriket and frequency and you will have the same tension between the test sprockets as you had on the engine. Place the blue belt on the fixture and apply the same load as used for the black belt and measure with the Kriket and frequency.

You should now have enough data to determine any difference between the black and blue belts under static load and static tension. You also have the measurement of the belt on the engine from both the original method to tension the belt and the later-introduced frequency method, both under static load and tension.

You are right that correct belt tension for the blue belt as compared to the black and correct tension for the earlier cars using the frequency method has been beaten to death. I got tired of all the opinions and guessing with absolutely nothing actually done to measure something. After consulting with both a physicist and a mechanical engineer I chose to make this fixture and actually measure something.

It is not impossible to come up with an accurate way to convert from one belt to another with actual comparisons. It is not impossible to convert from the old-style method to tension the belt to the frequency method. If somebody knows how it is done then they should post rather than pontificate. Direct comparison and elimination of all but one variable provides actual data. If this data is rubbish, don't use it.

I am simply offering a method for you to use to obtain the original objective.

 

[DailyDriver]

Note to SNOWRX.

In the Lotus service manual, Section EF, p 18a, under "Belt Tension Clavis Gauge" it says:

"In order for these benefits to be applied to the Lotus 4-cylinder 912/910/920 engines, the following belt tensioning procedure should be used on all such engines equiped with the 'HTD' (round tooth profile) type cam belt."

So, if you are using a trapezoidal belt it appears you are stuck in the old days with the old procedure and old equipment unless there is some way to switch to a round tooth belt. For any 912/910/920 engine with the 'HTD' belt Lotus has recommended a new, more "benefiicial" method of the Clavis Gauge.