Thank you for your comments.

Nothing wrong with the tools in this case. The operator is blaming himself - but trying to find the definitive cause. The cause is, in my opinon, and of others, (it can't be anything else really) a slack belt.

Now when it was fitted at 41Hz it seemed slightly slacker than it should be with the twist test - but not so bad it could be twisted as much as 90 degrees which some say is Ok (I don't - I say 45 degrees) so it is still a mute point.

I am 'old' school and admit I am prejudiced against rubber bands in engines. Duplex or triplex chain is still the best engineered solution to driving camshafts IMHO - they don't break (unless catastrophic failure elsewhere), and give plenty of warning if worn as they become noisy. I know modern toothed belts are an highly engineered item; simple, very good with better timing control, but suspect they are used because they are much cheaper to the manufacturer - no sprockets, no chain -all costly.

-------------------

The standard belt tensioner instrument is produced by Clavis. Which reads in Hz. There is an interesting article by them all about the principles: here:

www.clavis.co.uk/Handheld%20meter%20catalogue.pdf

QUOTE: Theory of operation:
The Clavis belt tension meter measures the natural frequency of vibration of a belt
span. This frequency is directly related to the tension in the belt. As the tension in
the belt is increased the frequency of vibration also increases.
The relationship between the measured frequency and the tension of the belt should
be determined from a calibration test on the belt span. A Clavis calibration rig is
available for those users of this instrument who wish to perform their own
calibrations, or we can undertake your calibration in our laboratory. Alternatively the
relationship between belt tension (T) and frequency of vibration (f) may be
calculated from knowing the mass per unit length of the belt (m), and the belt span
(l), using the expression;

T=4m l^2 f^2

Where T is in Newtons, F in Hertz, I in metres, and m in Kg/metre. However as the
belt does have some flexural stiffness the predicted tension for a given frequency
will be slightly greater than the actual tension. This is most noticeable on short belt
spans where the belt bending stiffness is the greatest. For belt spans greater than
250mm a calculation based on the above expression will provide results within 10%
of the actual values. END OF QUOTE.

In this pdf file belt parameters are given for mass per unit length etc but nowhere is there mention of "SEEM" units.

So who has invented these wonderful units? and what exactly are they?

No one here SEEMs to know ! :-)

Douglas Denny

Thank you FP for your continued help with this trying and highly irritating event.

I do not speak French but your finding that reference is helpful and makes sense. Typical of the French to invent their own obscure unit: probably to keep it exclusive to their engineers. Cynical moi? - never.

All I need now is to find the relationship between SEEM unit and resonant frequency in Hz - if there _is_ a relationship that is. Or, find the Peugeot official value of tension in Hz. Can anyone tell me that?

I agree with you the twist test between camshaft pulley and Diesel pump pulley is simple and more or less foolpoof. This fool relied on technology however, but to be fair to myself I thought it passed muster with a twist test that seemed slightly slacker than I would normally allow, but not too bad so as to be worrisome. The twist was between 45 degrees and 90 and the up/down movement slightly flabbier than I would normally allow, so tension definitely on the lower side but not silly.

Another factor might be: The engine did have a harsh knock on one cylinder until warmed up before the top-end overhaul was done; and after, on first starting up was knocking on all four more than I would have liked, but I put that down to a close tolerance fit of cylinder head with the new gasket and higher compression ratio than before on all cylinders. I am beginning to think now that it was too close a tolerance and the next thickness higher head gasket might have been wise rather than the minimum allowance as per the book.

Possibly the gasket tolerance was so tight, with the pistons almost touching the cylinder head, (it turned over no problem of course after fitting the cam belt), and the very slight slackness of cambelt allowed just a miniscule touch of a valve - which slackened the belt more, which then touched more with the valve, making the belt jump a tooth - then that was that - rippling down the valves as the crank turned. Probably all in a single revolution. That's my theory anyway. Ho Hum! - life's like that sometimes.

Douglas Denny.

Hello FP.

I have _had_ to move on - no choice - found another engine and am fitting it. Mistake has cost me about £300 in parts and another second hand engine. As you say learn and at least check with the 'normal' twist method next time as a safety check and do not allow myself to go slacker than the 45 degree twist criterion.

However good news: I have I think found final definitive answers now with some research on the internet.

Here is the Rosetta Stone of different units and data on cambelt tension measurement:-

www.autodata-online.com/uk/tbinfo.pdf

This has on page 9 of 11 a table of eight tension measurement devices and their units used. It therefore gives comparison of the units including SEEM units. It does not give a comparison of resonant frequency however. That has to be calculated; as follows:

Looking at the table: Lo and behold we have two SEEMS devices mentioned; a SEEM C.Tronic G2 105.5 and a SEEM C.Tronic 87 (curiously, and oddly, measuring slightly differently for a given tension value of force in Newtons for the first instrument which is a Burroughs BT33-73F / BT33-86J).

Here is the table comparison for the Burroughs (as above) and the SEEM C.Tronic G2 105.5:-

150 Newtons = 33 SEEM units

200 N = 40 SEEM

250 N = 47 SEEM

300 N = 53 SEEM

350 N = 61 SEEM

400 N = 67 SEEM

450 N = 72 SEEM

500 N = 78 SEEM

You will note the difference between SEEM units for each 50 Newtons increase is non-linear. These SEEM units are very strange indeed. Clearly they are completely arbitrary to the device.

So the next problem was to establish what the 41 SEEMs recommended in Haynes for the DW8 engine is in Hz. We proceed as follows:

Measuring the belt total length it was 1338mm = 1,338Metres. (It is marked at 1.4Metres, but we really need the mid-point of the belt section so I used the measured value as above). Measuring the span across the two pulleys (camshaft and diesel pump) which is what vibrates, was 240mm = 0.24Metres. Measuring the total mass of the belt was found to be 169 Grams for the new belt; and 166 Grams for the old belt. 169 Grams was used which = 0.169Kg. Hence the mass per unit length = m = 0.168 / 1.4 which = 0.12 Kg per Metre. We now have all the data for using the Clavis formula for frequency of a span of belt which is:-

Tension in Newtons = T = 4 m L^2 F^2

or, F = square root of { T / ( 4 m L^2 ) }

where F= frequency in Hz and L = span length in Metres. m = mass per unit length

As 40 SEEM units = 200 Newtons,

putting 200 for T (Tension) and finding F in the formula, Frequency = 85.1Hz

putting 250 Newtons for T, Frequency = 95Hz

-------------------

Conversely; as I set the belt at 41 Hz I can now find the tension in the belt which I set it to. Solving for 41 Hz gives a belt tension of 46.5 Newtons. So there was tension in the belt but it was only a quarter of what it should have been.

I measured the cambelt Hz for the replacement DW8 engine and it was found to be 68 Hz, that solves for a belt Tension of 128 Newtons; which is 0.68 (nearly three quarters) of the 200 Newtons it should be. I shall tighten it to 85Hz = 200 Newtons.

I think I am now satisfied I have explored this subject to its conclusion. Which is: yes I was in error. I paid the penalty. No the belt was not totally slack it was a quarter slack.

I also think the piston to cyl head gap distance was probably marginal and contributed to the problem however, with the absolute minimum head gasket thickness which I allowed for, and should have chosen one value thickness more.

One lives and learns.

Thank you all who have helped me towards a better understanding of what it is all about.

Douglas Denny.