Rally Polo MK4 S1600 - Intake Manifold Rebuild

[Benutzername]

It is extremely difficult to obtain technical drawings, documentation or detailed photos of the Polo S1600 engine. Getting hold of parts is virtually impossible! Nevertheless, I had a bit of luck and came across the original plenum chamber for the intake manifold quite by chance. As I was already working on optimising the engine of my GTI, this was a good time to rebuild an S1600 intake manifold. 

About the S1600 engine:
Depending on the documentation, different performance specifications for the engine can be found. These range from 215 hp to 232 hp and speeds of up to 9750 rpm. 
The 1.6-litre base engine has an aluminium block and, unlike the AJV, ARC and AVY, the connecting rods (Pankl) are 144 mm long instead of 138 mm. According to the homologation, the compression ratio was 13:1. This was achieved by using forged aluminium pistons. The crankshaft was standard cast iron. The flywheel (2976 g) was a special CrMo4 for a 184 mm Tilton clutch. There were apparently several camshaft profiles made from Schrick blanks. The valve train was rigid. In order to adjust the valve clearance mechanically, measurements were taken from the outside through 16 holes for each valve/rocker arm. This extremely complex process was necessary in order to achieve such high revs. 
In addition minimum two variants of the exhaust manifold, there were also several variants of the intake manifold, with only external changes to the plenum. The throttle body was mechanical and had an internal diameter of 56 mm. 

Let's get started then:

That was what I got. The complete plenum in brand new condition and this thing is really huge. But that makes sense – a very large volume of air in front of the intake runners.     

I still had an old original intake manifold from the AVY. So I simply cut off the flange and machined it on the milling machine to get started. 

I planned the runners with thick-walled aluminium pipes. Therefore, I milled a corresponding step into the flange so that it could be welded more easily later on. The flange was then adjusted directly to the cylinder head, which I was working on anyway. 

I have already roughly shaped the aluminium tubes and loosely assembled the whole thing on a cylinder head to get an impression of the size. I have the original lengths/dimensions and a bunch of detailed photos of the real S1600 intake manifold. That helped enormously, of course. 

 

 

 

 

 

 

 

[mk2]

Keep the radii nice and large to maintain a good laminar flow... Looking very nice

I read a great book years ago by a guy called David Vizard. The 'god' of tuning. He did so many experiments, just to get an extra 0.5 hp each time. Crazy, but altogether his modifications to a standard engine almost doubled the power. Very impressive.

 

[ObjectiveAway]

Very nice, thanks for sharing 👍

All FIA homologation forms are available online — I love looking through them, true engineering masterpieces. Pages 38 to 68 cover the Polo S1600 homologation form.

 

 

[Benutzername]

Then let's take a look at the next steps:

Base Plate and Trumpets:

It is, of course, difficult to obtain intake trumpets in exactly the same shape as the original ones. However, I found some at KMS that were at least the right length and inner diameter. They had a larger outer diameter, which could prove problematic later on if they did not fit through the large opening in the plenum. But first things first. 

After the trumpets arrived, I had all the necessary measurements to draw a base plate for the plenum, onto which the trumpets would fit on one side and the aluminium pipes could be inserted on the other for easier welding. The CNC milling machine did the rest.

Because the outer diameters of the ftrumpets were larger than the distance between cylinders, they had to be machined slightly. Now the trumpets fit into each other, which looks really cool and does not interfere with their function. 

After cutting the aluminium tubes to the correct length, the whole thing looked like this:

Edited September 24, 2025 by Benutzername

[Benutzername]

Fuel Rail:

The cylinder spacing of 82 mm is relatively rare in the world of engines. This becomes a problem when looking for a fuel rail. In the original part from AVY etc., the individual injectiors are slightly closer together because the intake runners also converge slightly. So something individual had to be created here too. For this purpose, I acquired a fuel rail blank. It is made entirely of aluminium and has a continuous 15.9 mm bore. I had planned to use 380cc Bosch EV14 (0280158051) injectors for the S1600 intake manifold. Because this type apparently has a very wide scatter band, I ordered matched injectors from from a swedish tuning shop. 

The dimensions for the injector seats were taken from the Bosch injector data sheets and milled on the CNC. M18x1.5 threads were cut at both ends for M18x1.5 to Dash6 adapters. I found a nice compact holder for these standard Bosch FPR on the internet that fit on Dash6. A 3 bar FPR is installed for now. That should be sufficient with the 380cc injectors. According to the Swedish report, it was only 368cc. Regardless...it's sufficient!

For the intake runner side, I made 4 sleeves and cut them to the desired angle. Since the injectors are actually intended for the US market, they unfortunately have a USCAR connector. So I needed adapters to EV1 plug so that I wouldn't have to convert the wiring loom. 

After welding, it took quite a bit of work to grind everything into the desired shape. Especially the insides...

 

 

[Benutzername]

Final machining and assembly

First, I ground the weld seams around the outside of the seats for the injectiors. From the outside, it was ready because heat protection was going to be wrapped around it anyway. 

The insides of the intake runners took a bit more time because you can hardly see anything with the straight grinder in the pipe. In the end, however, the result is quite acceptable. It looks a bit rough in the photos, but it is completely smooth (80-grit sanded), the four pipes are almost identical in shape, and the flow velocity in the channels should increase by 17% in purely mathematical terms. In that respect, I was satisfied. 

The final grinding work concerned the plenum. Originally, the flange was designed for a 56 mm throttle body. The flange had already been ground to 58 mm on the inside. So I adapted the whole thing to my 60 mm Bosch throttle to ensure a perfectly smooth transition. 

Nowadays, a relatively thin gold foil is often seen on racing engines with ITB as heat protection. I have my doubts about this... if it doesn't help, it at least looks important.  So I wrapped the pipes with this special heat protection foil and then started to assemble the entire manifold. Unfortunately, the original MagnetiMarelli pressure/temperature sensor does not fit, so I had to buy a suitable one from Bosch. I was even able to get exactly the same type of sensor that was originally installed. The Rally Polo didn't have a power brake unit, so I adapted the existing M12 hole to a silicone hose and use that for it. I had to cut the seals out of 1 mm sealing paper, as none of the original ones fit anymore. 

 

The replica of the S1600 intake manifold was now ready for installation.

 

[mk2]

Cool. Have you calculated the resonance rpm? i'd guess about 3500?

I LOVE the gold foil! 🤣😂👍🏻 Get a NASA sticker.

[Benutzername]

3 hours ago, mk2 said:

Cool. Have you calculated the resonance rpm? i'd guess about 3500?

No, it is higher. 3rd order is at 7350 1/min and 4th order at 5250 1/min. But I am curious to see what the fuel maps show later. Normally, things become clearer there. Unfortunetely, I am not quite there yet. I just need a bracket as support and then it will be on the road. 

 

A friend also said it looks like a little satellite. 🤣

Edited September 24, 2025 by Benutzername

[mk2]

Lupolink... 👍🏻

[ObjectiveAway]

Excellent result, a pleasure to watch! 🤩 Usually in such processes, the unpleasant aspect is that everything takes a very long time — how long did it take you to make such an intake manifold?

[Benutzername]

1 hour ago, ObjectiveAway said:

Excellent result, a pleasure to watch! 🤩 Usually in such processes, the unpleasant aspect is that everything takes a very long time — how long did it take you to make such an intake manifold?

Thank you very much! 👍

I received the plenum in the last week of July. Then I got started pretty quickly. On Sept 1st, the welding was done, and the following day, the seats for the injectors were welded. On September 9th, it was sanded and wrapped in foil. Then there was a short break because I had to go on holiday with my wife...🤐  Final assembly took place on Sept 17th and the installation with engine start-up on the same day. 
If I exclude the holiday, it took 6 weeks. 

[Benutzername]

I thought long and hard about how I could find a sensible solution for the air filter. My initial thoughts were to use existing OEM air filter boxes, but there isn't much space, and such small boxes usually only have a small cross-section. So why not 3D print it myself? While I was looking for large 3D printers, it occurred to me to print the entire intake manifold. I still have a space problem because the intake manifold actually has to be slightly curved. 

So I ordered a large 3D printer (420 mm x 420 mm x 500 mm print volume) and started designing. First, I needed a perfect match for the inlet flange.  

Once that was sorted, I drew the entire manifold in CAD and did various test prints. I had imagined it would be more complicated, but the quality improved from prototype to prototype. So I printed a first curved pipe and a trumpet to check the installation space. The cross-sections are perfect, but the curve was too sharp and it was too long, so I shortened it again and adjusted the curve. But in the end, it's an iterative process... 
Here are a few impressions of the first parts:

Last night, I printed another version, more on that later. First, I have to dismantle the plenum. “Someone” has sealed it super tight with silicone...

The goal is to either print the entire part from PA12CF20 myself or to use the prototypes for installation space testing and then have the final parts printed in aluminum by a company.I am printing my current prototypes with PETG. This is not temperature-stable enough, and gasoline would make the material brittle.  

 

Let's see where this goes...

Edited December 15, 2025 by Benutzername

[Benutzername]

This is the latest version installed on the plenum. It looks good so far. I'm curious to see how it fits in the engine compartment. 

[Benutzername]

Better than expected! The long version already fits perfectly. Long version means original lengths like the S1600 intake manifold. .

[Benutzername]

Once everything fit properly, I moved the injector seats as far away from the cylinder head as possible, drew in two bosses as fuel rail holders, and finally added a nice stiffener to the base plate with a honeycomb pattern.

At the same time, I am already doing various test prints in the hope of soon being able to print the complete and solid intake manifold reinforced with carbon fiber. 

[Benutzername]

After a few trials with PA12CF, I got really good results. So I started printing the entire component using these settings. Printing took over 50 hours. The finished part turned out impressively well. It still needs to be annealed to achieve its final heat resistance and strength. We have a good oven for this at the company. 😉 

[Benutzername]

The annealing process went well. It did not cause any further warping within the component. The base plate is slightly warped, but this can be straightened by grinding. 

The Trumpets were then subjected to the same process.

For the fuel rail, I glued in M5 threaded inserts and small M6 stud bolts for the trumpets. Now I could assemble the part and enjoy the result. Everything fits perfectly!

I still need the right gaskets, and some surfaces still need to be sanded down. I also need to mill a heat decoupling in the form of a 5mm Pertinax plate. Although PA12CF can be used up to 175°C, the material starts to creep at around 110°C. In the worst case, this could lead to leaks, even though the temperatures at the cylinder head would never exceed 100°C. 

Edited December 18, 2025 by Benutzername

[Benutzername]

I actually wanted to design and print more trumpet variants, but the tubes also need a few holders so that everything is well organised. 😁

 

[Benutzername]

Unfortunately, I am still waiting for the Pertinax plate to be manufactured. Our CNC milling machine needed a minor repair. Unfortunately, nothing can happen until that is done. 

Between Christmas and New Year, I took care of getting a useful set of trumpets. These now range from 40 mm to 70 mm in length in 10 mm increments. Anything shorter than 40 mm would hardly make sense given the desired speed range, and anything longer than 70 mm would be critical in terms of space in front of the trumpet inlet within the plenum.

I will start my tests with the 60mm version. Here is a comparison with the original aluminum version. 

 

 

Let's move on to the next stage of evolution: the channel in the intake pipe is tapered from around 42 mm to an oval shape measuring 47 mm in width. This results in an undercut. This can be easily changed by also designing the round inlet as oval. The oval in the inlet must be slightly wider than at the flange to the head, but the cross-sectional area must be the same as the 42 mm round.

After calculating the needed crosssections, I redesigned the entire intake manifold. In CAD, the trumpets now look extremely huge, which made me a little suspicious. So I test-printed a set of trumpets.

I measured and recalculated everything again – it fits! Good, then it stays that way. I will also print the EVO2 manifold in PA12CF and test it later. I don't expect any major improvements, but every little bit helps, right? And even if it doesn't work, it's still a nice activity and looks good on the shelf... 😆

 

[mk2]

Please make it variable length, based on RPM. Just use a servo or stepper motor.

[Benutzername]

3 hours ago, mk2 said:

Please make it variable length, based on RPM. Just use a servo or stepper motor.

I've actually already thought about that. It should also be feasible in terms of space, with a length change of approx. 40 mm. I also have small servos at home that could be controlled with the Haltech via generic maps based on load and speed. I like these kinds of gadgets, but I would first test the shortest against the longest version of my trumpets. If this results in the potential to achieve a wider speed range, I will give it a try—EVO3 maybe?

[mk2]

I don't know. The tricky part will be matching the length to the variable cam... And the greatest effect will obviously be at WOT.

Looking forward to your experiments!  👍🏻