[Julian Klepacki]

Paul, you are 100% right about the lower seal positioning, as I found out after attentively glueing it all in place.

Only to see after much faffing with the glass and upper seal and concluding it was obstructing optimal fitment of the glass. I saw the seal touching the glass more on the ‘mound’ of the seal. I also had seal extending beyond the edge of the glass. Looked wrong. Also the lip that sits on the parcel shelf didn’t look right. Like it wouldn’t be pressing down enough on the shelf?

When i glued it on first time, i kinda let the seal sit where it placed itself naturally on that rolled upper-edge of the parcel shelf. But, then after above figured/concluded it needed to sit more inward on that rolled-edge. This would effectively position the seal-lips higher up and move the mound of teh seal inwards, away from the glass.

So I had to pull the glass again, and pull off the seal and glue it all back slightly inward on the ‘arc’ of teh rolled-edge. Yeah, it made a difference. I used the edge of teh glass as my reference and positioned the seal so the lip edge was flush with it. Again, not recalling what it was like before removal of everyhitng. This seeemd to be the right ‘finish’ for it?

Now you can see where the glass is touching the seal along that edge, much better.
Yeah, i expect some fun and games with sealant whilst attempting to get it all water tight?
It was water tight in that area before, as i never recall ever having ingress issues in that area?
Lets hope it can be again.

[Julian Klepacki]

i had to read up on pva release agent, as to me, pva was wood glue or stuff added to plaster/cement to aid bonding etc. Alcohol as opposed to Acetate?

I read pva lends itself where other agents are suitable, but are others an option; wax, sprays etc?
Good luck with plan B.

[Julian Klepacki]

i didn’t get what was going on here at first :/. But with pictures i geddit.
Ingenuous. So, is the plan to pop the matt out, then cover it with whatever?
Looking forward to seeing the results/progress.

[Julian Klepacki]

So, i seemed to have sorted my poorly fitting rear screen. I took the replacement out and compared it to the OEM i removed. Measurement wise it was the same or within 5mm, mostly the same. Those 5mm variances i can easily attribute to measurement error. I used like a tailors measure and having to manipulate around the compound-curves could easily introduce a few more mill here and there?

Anyway, I also compared physically, positioning them opposite to like give a mirror image in attempt to compare the curves. I’d say they were. Considering the relative simple way i could do it and trepidation of breaking one/them. In particular the new replacement. I concluded they were if not the same, a very very close copy. more than good enough. Confidence repros are good enough. On the rear anyway.

I then compared the upper-seal. They looked the same, profile, material and everything. I’d say the better nick used replacement i was fitting was OEM too? I went with that one again, but i trimmed a little off each end that over-hung the screen at the each end as i thouht this also could be contributing to issues. i left about 7mm extending, whereas before it was more like 15-20mm before.

one thing i noticed and previously concluded what the problem was that the screen seemed to be forced over to the near-side somewhat. This seemed to be done by the lower-seal? Looking at the car with everyhting removed, you can see a difference in the installation of parcel shelf. there is less gap between shelf moulding and bottom of C-pillar inner on the NS, shifting the screen over to that side? Hope that makes sense. It’s never been replaced and conclude it’s simply from build. They are quite poorly made afterall aren’t they?

So this time, i pulled out the lower-seal at the sides so the screen was free sit central without any side-to-side interference. It seems to have done it. It’s a much closer, tighter fit with only slight movement of seal at the sides, whereas before there was a lot. I’ll have to just push/poke the seal into place on those sides. I conclude it can’t be any better given the position of the lower brackets forcing the screen up & down? In fact with added additional ~3mm rubber strips to those brackets to place the glass just that bit further up & down. unfortunately i have no reference as to how ‘snug’ the seals were before i pulled it all apart?

One thing is I have a small gap between the inner upper seal and the parcel moulding. See below. These were also with the original installation, as i had to remove some mastic like material that had been placed in there? I will add some mastic/sealant there in attempt to seal it up there.

Off-side

Near-side

One thing with the repro screen is that the heated-element isn’t quite central on the screen. It’s about 10mm closer to the edge on the NS. The OEM screen was equal mesurements. This results in unequal visible positioning of it. I can live with that.

NS

OS

i’ve now given the floor areas a couple of coats of jotum 90, with some treatment in rusty areas with Hydrate-80 first. Awaiting delivery of a box of Silencoat butyl soundproofing pads, as i’ve used on the inner door skins and inner rear-wings.

i need address a missing captive nut for the drivers-side seat-base. it’s broken out. i will weld in a small plate with a nut welded to the back and touch up the jotum.

One thing that concerns me is that the accelerator linkage appears to have come undone? When i removed the panels around that area to gain access of the floor. I was careful to try and leave it undisturbed. But with having to move it now and then, it doesn’t feel right now and it looks to have come out from the hole in the NS g.box bulkhead?

Concluding it had disconnected i moved it right out the way for maximum access. I hope this is somehting i can easily reconnect? I recall a thread on here someone replacing a throttle cable having a nightmare with it? Any advice here welcome.

[Julian Klepacki]

i’m gonna get some of each and build up some spares.
even without the proper bodies, worth getting a compatible body, like that lucas part and make something useable. A must to have in the spares when travelling abroad or anywhere really in my book?

i’m lucky when i bouht my car it came with a lot of spares, a lot.
injection wise; fuel pumps, MAP sensor, throttle sensor and complete analogue control unit.

That reminds me i have circuit diagrams of the control unit and some other info about the pcb, identifying what areas control what. I’ll see if i can attach them to a post or email them to you if you’d liked them?

They were emailed to me from a guy called stuart, back in 2011. I suspect he is/was a club member and was compiling a infodoc on the injection system.

[Julian Klepacki]

200 ohm NTC Thermistors aren’ so common, but Mouser have a reasonable selection available.
would be worth getting and making up a spare air-sensor too?

https://www.mouser.co.uk/c/circuit-protection/thermistors/ntc-thermistors/?resistance=200%20Ohms&srsltid=AfmBOoqgNCNCD9QEsfFekRqefV4KejGSzjczZINm-DpvRVbcuUovgvx-

Simon, what/whose/which 2k NTC did you end up buying?

[Julian Klepacki]

Where i removed the under-padding from the front section of loor. It got destroyed in the removal.

Anyway, i have my jotun now, so i’ll consider my options and my direction with padding/sound-proofing whilst i complete the floor painting.

regarding the stitching, i get it. One thought though, stitching the citroen way, i’d think the second run of stitching from the upper-side throuh to the loose binding folded under is at risk of moving and messing up? Guess thats where the walking-foot comes into play?

Dunno, i’d be inclided to stitch the binding to the upper-side, fold over to the under side and then stitch from the under-side, keeping the binding taught as fed through the machine? using the foot as a guide at the folded-edge to keep parallel, plus shift the needle over to get a broader seam? Just thoughts? A practice of methods on some scrap carpet beforehand will likely dictate which way to go?

lots of options.

[Julian Klepacki]

ARB – anti roll bar? i geddit? cover is good, i do plan on refitting it.

understand about the carpet stiching. initially i thought, how can you stich the underside, without it becomingn visible on the top side. But you are doing it away from the edging, so it’s kinda camouflaged within the carpet. Providing thread is same colour as carpet? Or as pauls describes; stick through the edging again, but you only see one line of stiching for a neater finish, providing it’s parallel etc?

I may even already have a walking-foot that came with sewing machine, have to check.

Paul, did you get replacement cotton/hair padding for the floor pastie or manage to reuse the old stuff?
I may go for replicating it, but possibly use an updated material for the filling? Thinking the Plastozote closed-cell business?

i’ll ponder this lot whilst jotunning areas and messing with the rear screen.

Peter, i considered soldering the rivets, but though inevitably the plastic would start melting trying to get enough heat into it all? Drilling rivets out was very easy, a 3mm drill, spun just enough to take the rolled-edge off. i wanted brass M2.5, but could only get hold of zinc in time. It worked out well and easily dismantable as required?

Re-silvering the inners would be good, but don’t think any paint is avaialbe that doesn’t end up just like silver colour paint? I have seen some witchcraftesque ‘chrome’ affect paint on the interweb, but unsure of it’s reality? I have carefully stuck tin-foil into inners before with acceptable results?

[Julian Klepacki]

super! i don’t think that could have worked out better?

[Julian Klepacki]

Thanks for all the input.

Paul, regarding rear screen. No repairs to C-panels etc. But you made a good point on the upper-seal. I have changed it to one that looked in better nick than the original. It looks OEM, But it may not be? I’m going to remove it all, close compare, measure etc etc.

I did fit the original screen with the replaced upper seal and after seating the very lower-end of teh seal properly it fitted OK. I don’t have any strong memory or concerns about it, so conclude it was as expected?

Simon, i’m going to replicate what you did for the boot. I had seen your post before, but it slipped my mind. Exactly what i’m after. Can you expand on the stitching of the edging? I get stiching the edge, folding it over to hide the stiching, but how is the other edge stiched without it being visible on the other side? Is it glued?

Also what is the ARB & a Transome?

Regarding floor/underlays. Lots of ideas. I’ve not used jotun before, and nearly out of POR, i’m going to get some to try out. Reading about it ’90’ has replaced ’87’ and possibly the ‘Aluminium’ or ‘Glass’ refinforced stuff gives added protection/durability.

What about thinners/activator for it? Does it have to be the Jotun 17 stuff or can others be used?

i’ve got numerous sheets left of sound-deadening panels. The sticky-backet, foil-lined, tar/bitumeny type stuff. It all seems to be the same stuff whoever bought from; NK, DD etc? So i’ll just use that up.

Interesting to see in Paul’s photo the bulkhead & that cross-section carpet is sealed under the edging of the floor-covering.

Regarding the rear lights. Those bits shown above are the spring-contacts for the ground on the lamps & the links connecting the contacts together. Due to all being oxidised I had very high resistances and in a couple of cases open-circuit.

I simply drilled out the little brass rivets, cleaned everything up to clean metal, applied a little contact-grease to mating surfaces and used M2.5 machine screws to connect it all back up. I now have stable 0.5ohms measured between lamp ground contact and ground-lead.

the rubber cups were somewhat deteriorated and damaged. So i got some silicon end-caps off ebay. punched a little hole through the centre and they worked out ideal?

[Julian Klepacki]

Simon, yeah, that thermal compound/grease is mostly the same. like a white grease you see between semiconductors & heatsinks etc. SOunds like you are sorted there.

i think most of us are cut from the same cloth with regards to having to dick about with stuff. i bet as a child you took your toys apart to see how they worked soon after the initial novelty of playing with them wore off? i know i did?

anyway, look forward to seeing results of your endeavours.

[Julian Klepacki]

Thats as good as perfect? All you want is the main body really?

When i have done it before, not for a DS, but motorcycle oil-tank, in principle the same. i’ve filled the lower cylinder part with heat-sink-compound using a small 5ml syringe, helps keep contamination to a minimum from the upper part to get the best adhesion when you come to seal it.

you can get epoxy with thermal conduction properties, for potting electronics. 3M & Electrolube make it. But it’s not practicable, unless you are thinking of going into production and becoming a DS temp-sensor oligarch?

prep the bead/sensor; trimm the leads back and soldered a pair of tails for connection. slip some heatshrink over the tails, at least one if room is tight?

Then another sleeve of heatshrink over the pair of wires for strain-relief. At least an inch or so protruding from the body for some support or longer if a 2-core lead is preferred?

plunge the lot into the body, burying the sensor into the heat-compound. clean up if required around the upper internal. i used cotton-buds with acetone.

seal it up with something appropriate. i used silicon as i thought it would introduce some flex for the wires exiting the body. Also good seal and heat/contaminate friendly?

maybe you can re-use the top-cover, fill that with silicon too, wires coming out where the blades protruded or make a hole for the 2-core? let excess silicon ooze out for a proper seal?

if/when you get round to it, i can send you a syringe + heatsink compound, saves you buying? just let me know.

[Julian Klepacki]

Simon,
Thats fantastic outcome.
That Lucas unit looks to be a 2K NTC, as tables give a little over 2K4 at 20C & i suspect your temp is not a calibrated source, so ~2K5 @ ~20C looks cock-on?

apart from personal satisfaction, opportunity for tinkering, is it worth bothering trying to rebuild/repair an oem unit? if it doesn’t look odd with that 19mm hex in situ or your car isn’t concourse better than factory. does it matter?

i guess there is scope to machine down that hex as small as possible, maybe get close to 13mm AF?

I see no reason why a 2k2 ntc wouldn’t work, it’d be close enough?
But 2k ones are easily obtainable, so can make one as OEM?
mouser have lots of them.

https://www.mouser.co.uk/c/sensors/temperature-sensors/?product%20type=NTC%20Thermistors&resistance=2%20kOhms&srsltid=AfmBOoqn59vUPluJn3Cdlm6k6V1_TKM57HUEhy-zjai83h0cONVVXegG

Farnell have them too:

https://uk.farnell.com/c/sensors-transducers/sensors/temperature-sensors-transducers/thermistors/ntc-thermistors?resistance-25degc-=2kohm

The air-temp looks to be a 250 Ohm NTC, not so common, but out there.
link to temp/resistance charts, section 13.4.3:

https://jetronic.org/index.php/en/d-jetronic/72-troubleshooting.

your air-temp sens looks to be OK.

Oh, i also found out the NTC ‘II’ refers to the ‘second’ NTC, i.e. engine temp.
NTC ‘I’, is the other one, the air-temp sensor. nothing do with accuracy :/.

OK, thats my tea break over, back to working on the car (ds).
glad you got it sorted and identified an excellent source of alternative part, cheap as chips too.

[Julian Klepacki]

oh, let me add, the OEM sensor is simply a bead NTC encapsulated into an appropriate housing. So in theroy you should be able to make one as good as OEM.

In fact, if they are so difficult to find and expensive, i need to make some for myself?

Also, the sensor causes an elongated injector-pulse to give more fuel. So, as you say the higher-resistance equates to a perceived lower-temp resulting in longer-pulse and too rich mixture at normal temps, leading to the misfires and i guess eventual sooting up of plugs.

[Julian Klepacki]

i have some info on these, you are probably aware, but a lot of useful info is here:

https://jetronic.org/en/d-jetronic

I know it is a ‘thermistor’ of sorts, but they come in various ‘standard’ values based on the 20C temp, i.e. 5k, 10k, 25k etc.

i’ve dug out my bosch jetronic book to see what it had to say and in there it states it is a ‘NTC’ Thermistor, and refers to it as a NTC ‘II’. The ‘II’ almost definititely refers to the accuracy; 1%, 5% etc, But unsure of details, not overly tight accuracy by the looks of it.

lookign at the curve i have from info on the link given above it deviates more the lower the temperature:
10C 3k to 4k5, centre temp 3k7
20C 2k1 to 3k1, centre temp 2k4
30C 1k4 to 2k1, centre temp 1k65
40C 1k to 1k4, centre temp 1k1

NTC tables imply it is a 2kohm NTC. From 2k NTC tables, centre temp:
10C 3k68
20C 2k44
30C 1k65
40C 1k14

So it looks like a 2K NTC.
Other ‘std’ NTCs will be significantly different: 3k, 4k7, 5k, 10k, 15k, 20k etc etc. So the 25C temp will be significantly off, yeah?

You can get very small ‘bead’ NTC’s for not much money from the likes of RS Components, Farnell, ebay, whatever.

I don’t think it’d be that difficult to make one in the absence of the real thing?
Certainly as an interim measure until you find OEM, if desired.

Buy a cheap sensor for the body, get two or three to experiemnt. Gouge out the guts, rebuild with new 2k bead NTC. I bet you could make one like OEM easy enough? Food for thought?
I have done similiar with resistance-senders to match up values for my own home-built guage(s)

if you do go diy route, get some heatsink compound to embed the senor into within the body for optimal heat transfer. it doesn’t cost much, again RS, ebay, whatever.

if you want photo of chart, tables etc let me know i’ll sort it out.
good luck with it.

[Author]

Posts