• Harrop Engineering doing the "impossible" - Introducing the world's first positive displacement E92/E90/E92 S65 V8 supercharger

      Wow. That is exactly what came out of my mouth when I saw this supercharger kit at Bimmerfest. In all honesty, it was my main motivation for attending the event once I heard it would be there. Some people may be reading this and wondering what the big deal is as superchargers already exist for the M3. That is true, superchargers do exist, but they are all centrifugal style blowers. This is the world's first positive displacement supercharger for the S65 V8.

      Why is that a big deal? Because it's a completely different type of blower with completely different power delivery. This is the solution for those who complain about the M3's torque or for those who do not like the response down low of a centrifgual blower which needs rpm to make boost. This style of supercharger is always making boost, from anywhere in the rev range.

      That is what makes incorporating it on the S66 V8 difficult with its independent throttle bodies as positive displacement blowers usually are set to pull air through a throttle body not blow into it. The centrifugal blowers send air into a manifold. If a positive displacement were to use a similar setup it would be blowing air in while the throttle bodies attempted to close as it is always making boost. See the problem?

      The solution Harrop Engineering came up with was a manifold that fits in between the throttle bodies with a bypass valve fit that relieves pressure as needed and makes sure boost is not made as throttles are closing. Pretty trick setup eh? You can see the design in the photos and also how tight the packaging is. I do not even want to know how many hours were put in trying to get this all to fit.

      The blower employed is a TVS1740 unit from Eaton. This is a roots blower and likely will be set to 5.5 psi or so initially. Boost from this blower is different from a centrifugal and will be harder on the stock rods since it is at full boost right away. Expect horsepower in the low 500's to the wheels.

      Oh and see that little black box to the left of the manifold off a metal area at the inlet? There will be electronic boost control which is likely the first of its kind on a positive displacement setup although I am not able to confirm this. Different maps with different boost levels for different fuel? A possibility, yes.

      This is a big deal and an engineering feat for the S65. My complete and utter respect and that of this network to the guys at Harrop. There is still work to be done here. It is not quite ready although pricing is initially set at $12,990. There is still a lot of tuning to be done but this will be coming eventually.

      Pictures below, much respect Harrop and it was a pleasure meeting you guys at Bimmerfest. You were incredibly nice and answered all my (MANY) questions. Thank you:























      This article was originally published in forum thread: Harrop Engineering Twin Screw S65 V8 Supercharger started by CookieCrisp View original post
      Comments 276 Comments
      1. DFM's Avatar
        DFM -
        Click here to enlarge Originally Posted by Sticky Click here to enlarge
        The adults will continue to discuss this and you are removed from the thread so it can stay on track as you are taking away from a great discussion.
        Lost marine is gone?! we were just getting started!!!!
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by DFM Click here to enlarge
        Lost marine is gone?! we were just getting started!!!!
        Guy is in over his head when he tries to discuss technical aspects it's been a repeat pattern for years. I do not believe he is very educated but it would be great if he did a little more reading and kept great threads like this on topic.

        Regardless, I'm still stoked on this setup and can't wait to see more.
      1. DFM's Avatar
        DFM -
        Well it'd be nice to have someone refute my post. I said a lot of technical stuff there and would like to know how accurate I am.
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by DFM Click here to enlarge
        Well it'd be nice to have someone refute my post. I said a lot of technical stuff there and would like to know how accurate I am.
        Well then hopefully someone else chimes in Click here to enlarge
      1. Jonathan@Aviva's Avatar
        Jonathan@Aviva -
        Hey DFM, Ill do my best to expand on some topics. Much respect implied and thank you for your contribution.

        Click here to enlarge Originally Posted by DFM Click here to enlarge
        Turbo lag cannot be entirely eliminated without the use of additional anti lag technology. Regardless of how close time to full boost is for each platform, it will always be longer for the turbo (especially in lower RPM's). The computer needs to be able to progressively adjust the timing and fuel as the boost rises at an exponential rate. With a PD blower, boost is similar to an on/off switch, so the progressive adjustment of fuel and timing is more immediate. Where did the 2500rpm for the PD come from? pretty certain a PD blower hits max boost much sooner than 2500 rpm...
        In theory a PD pump / blower has a fixed displacement per revolution, in the same basic theory an engine is the same type of "pump". The TVS is a roots type supercharger with NO internal compression ( as opposed to a screw with an internal pressure ratio and work * back to this later ). The supercharger like an engine has its own volumetric efficiency curve.
        ref:


        Boost response is instant. Sometimes you control / adjust the bypass mechanism with a damper, restrictor orifice or ( digitally damped ) to bring boost on softly. Though this is over fractions of a second. Time-to-torque from driver request to torque at the tire is still instant, even compared to a small turbo setup its drastically noticeable. It's literally instant tire shread.

        Boost curve. Pressure is not necessarily constant over the power band. Refer back to the volumetric map.
        A small supercharger ( relative ) matched to a motor will have a very flat boost profile, for example from 2000-8500 RPM. While a larger supercharge matched to the same motor, with the same peak boost pressure request, will have a more ramped plateau. 60% of peak boost down low, to 100% of peak boost by redline. This is not lag, its just the result of the interplay of rotor speed, supercharger volumetric eff vs rpm, and the engines vol eff. With this charger I'm confident that this will be a complimentary component to the character of the car.

        In summary:
        Boost Threshold : is not a factor with the PD system. Its always able to make positive pressure. While a turbocharger there is only so much potential energy for a certain pressure ratio after a specific RPM point.
        Boost profile : PDs have flat to slightly ramped ( size dependent ) profiles. Superchargers since they are directly coupled to crank speed, have a FIXED boost profile.
        Lag or time to torque : Instant torque, reaches boost threshold immediately. as fast as the bypass valve is allowed to snap shut, Lag is non existent.



        As a function of RPM, because the turbo can probably push more air at the higher RPM's than a PD blower at the same boost level. You're right, it depends wholly on the turbo itself. Think of a jet engine. A jet engine is very efficient because the airflow is in one constant direction. Now with a car, we want to emulate this strait through flow as much as possible. A boost adder of any kind will greatly affect the flow of air from air filter to exhaust tips. The differences in the overall flow are extremely different for turbos vs PD blowers. How exactly does this affect the tuning? well, I would think it would only really affect the tuning if you are tuning the cam timing. Cam timing affects scavenging, and scavenging is affected by the airflow through the engine.
        Cam timing for a PD system, can be quite complex. There is tremendous gains to be realized if you get it right. Especially intake cam phase. You can control boost pressure with cam timing.

        I disagree that its the same system theory. Turbo setups control boost by modifying the source of energy (exhaust gasses) and how that energy gets turned into boost. This PD blower controls boost by venting it. It has nothing to do with changing what is creating the boost in the first place, therefor the load the supercharger puts on the engine doesn't change, just the pressure in the manifold. Does this actualy affect the tuning? YES. Venting the exhaust gasses reduces the energy going into the turbo, which reduces the speed of the turbo, which modifies the flow volume/velocity of air at the compressor outlet, which lowers the boost in the manifold. Venting charge pressure (PD setup) does just that, vents the charge pressure and reduces boost very quickly. Now I can see how the E-BOV could be controlled similarly to a wastegate, but exactly how you control it is different because the wastegate is the start of a chain reaction in the piping, whereas the E-BOV is nothing of the sort.
        Correct:
        Turbo charges load the motor through pumping loses. Turbine inlet pressure which drives turbine shaft speed / boost, is back pressure. It is a very real pumping loss.

        The TVS / roots does no work internally. It does not compress the charge between the rotors. When it is in Bypass operation, it virtually has NO load, since no work is being done. Other PD blowers are always doing work always compressing, that is why some also have complex drive clutches ( Mercedes ). So if its not making boost, its not drawing power from the crank.


        confused on this statement, can you please explain a bit more thoroughly?
      1. benzy89's Avatar
        benzy89 -
        Click here to enlarge Originally Posted by Sticky Click here to enlarge
        Guy is in over his head when he tries to discuss technical aspects it's been a repeat pattern for years. I do not believe he is very educated but it would be great if he did a little more reading and kept great threads like this on topic.
        So instead you ban him from the thread, preventing him from getting any additional information (or potential knowledge) of this setup on the S65 Click here to enlarge
      1. Sledgehammer's Avatar
        Sledgehammer -
        So we just banned one person for asking legit questions... DFM and Lost had some good stuff going here without a bunch of sarcasm or ego being involved. Good job on that
      1. DFM's Avatar
        DFM -
        Click here to enlarge Originally Posted by Sledgehammer Click here to enlarge
        So we just banned one person for asking legit questions... DFM and Lost had some good stuff going here without a bunch of sarcasm or ego being involved. Good job on that
        Agreed.
      1. DFM's Avatar
        DFM -
        Thank you for taking the time to type this outClick here to enlarge

        Click here to enlarge Originally Posted by Jonathan@Aviva Click here to enlarge
        Hey DFM, Ill do my best to expand on some topics. Much respect implied and thank you for your contribution.



        In theory a PD pump / blower has a fixed displacement per revolution, in the same basic theory an engine is the same type of "pump". The TVS is a roots type supercharger with NO internal compression ( as opposed to a screw with an internal pressure ratio and work * back to this later ). The supercharger like an engine has its own volumetric efficiency curve.
        ref:


        Boost response is instant. Sometimes you control / adjust the bypass mechanism with a damper, restrictor orifice or ( digitally damped ) to bring boost on softly. Though this is over fractions of a second. Time-to-torque from driver request to torque at the tire is still instant, even compared to a small turbo setup its drastically noticeable. It's literally instant tire shread.

        Boost curve. Pressure is not necessarily constant over the power band. Refer back to the volumetric map.
        A small supercharger ( relative ) matched to a motor will have a very flat boost profile, for example from 2000-8500 RPM. While a larger supercharge matched to the same motor, with the same peak boost pressure request, will have a more ramped plateau. 60% of peak boost down low, to 100% of peak boost by redline. This is not lag, its just the result of the interplay of rotor speed, supercharger volumetric eff vs rpm, and the engines vol eff. With this charger I'm confident that this will be a complimentary component to the character of the car.

        In summary:
        Boost Threshold : is not a factor with the PD system. Its always able to make positive pressure. While a turbocharger there is only so much potential energy for a certain pressure ratio after a specific RPM point.
        Boost profile : PDs have flat to slightly ramped ( size dependent ) profiles. Superchargers since they are directly coupled to crank speed, have a FIXED boost profile.
        Lag or time to torque : Instant torque, reaches boost threshold immediately. as fast as the bypass valve is allowed to snap shut, Lag is non existent.
        This just confirms, to me at least, that tuning the two FI methods cant be compared with a number. the behavior of the boost curve for a turbo vs a PD blower (especially in the first few seconds after opening the throttle) isn't really comparable. The methods used to get the engine to run smoothly are different, since they depend on much different behaviors of the airflow through the engine. Excellent and easy to read explanation of PD boost response by the way, didn't leave me with many questions.





        Cam timing for a PD system, can be quite complex. There is tremendous gains to be realized if you get it right. Especially intake cam phase. You can control boost pressure with cam timing.
        I certainly don't know the specifics of cam timing for a PD system. Would be interested to know more about this. any good resources you know of?



        Correct:
        Turbo charges load the motor through pumping loses. Turbine inlet pressure which drives turbine shaft speed / boost, is back pressure. It is a very real pumping loss.

        The TVS / roots does no work internally. It does not compress the charge between the rotors. When it is in Bypass operation, it virtually has NO load, since no work is being done. Other PD blowers are always doing work always compressing, that is why some also have complex drive clutches ( Mercedes ). So if its not making boost, its not drawing power from the crank.
        Thank you for clarifying that for me. I should have known that because a friend of mine has a supercharged 4cyl Mercedes w/ a complicated drive clutch. Lets just say when you start changing the crank pulley to get more boost, the drive clutch is unhappy Click here to enlarge

        Since you have a good knowledge base on this subject, do you think you can just put a percentage number on how similar it is to tune a PD vs a Turbo on the S65?
      1. GuidoK's Avatar
        GuidoK -
        Click here to enlarge Originally Posted by Jonathan@Aviva Click here to enlarge
        The TVS / roots does no work internally. It does not compress the charge between the rotors. When it is in Bypass operation, it virtually has NO load, since no work is being done. Other PD blowers are always doing work always compressing, that is why some also have complex drive clutches ( Mercedes ). So if its not making boost, its not drawing power from the crank.
        This is imho not an accurate explanation. You make it beleve that other PD compressors (like the twinscrew) are compressing the air (internally) in a situation that it has nowhere to go (and thus costing a lot of energy and creating heat). That is not true. The air is 'squeezed' directly out the outlet (more like a peristaltic kind of way). It flows out immediately. It's compression with a hole (the outlet) at the end Click here to enlarge
        That is no different than taking a pocket of air and transporting it to the outlet (what a rootscompressor does). The efficiency of a bypassed twinscrew or roots/tvs is therefore about the same.
        Clutches are used because transferring air without building pressure also costs some energy as does the movement of the gears/lobes.
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by benzy89 Click here to enlarge
        So instead you ban him from the thread, preventing him from getting any additional information (or potential knowledge) of this setup on the S65 Click here to enlarge
        Sometimes it's necessary to keep a thread on course so EVERYONE ELSE can learn without distraction.
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by Sledgehammer Click here to enlarge
        So we just banned one person for asking legit questions... DFM and Lost had some good stuff going here without a bunch of sarcasm or ego being involved. Good job on that
        Thread viewing can be lifted it's quite simple I think it's best to let this thread stay on course with the great posts we are getting for now so it does not continue towards OT posts. Thanks for your concern.
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by DFM Click here to enlarge
        he behavior of the boost curve for a turbo vs a PD blower (especially in the first few seconds after opening the throttle) isn't really comparable.
        And hence leading to difficulty for the tune.. and the stock rods of course. So this setup is definitely quite different and requires different approaches/tuning.
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by Jonathan@Aviva Click here to enlarge
        Cam timing for a PD system, can be quite complex. There is tremendous gains to be realized if you get it right. Especially intake cam phase. You can control boost pressure with cam timing.
        So essentially somebody is doing some heavy VANOS work?
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by GuidoK Click here to enlarge
        The efficiency of a bypassed twinscrew or roots/tvs is therefore about the same.
        Based on...? The efficiency of a twin screw over a roots has been shown at high rpm or with high volumetric efficiency whereas a roots tends to have better efficiency at lower boost from every comparison I have seen.
      1. GuidoK's Avatar
        GuidoK -
        Click here to enlarge Originally Posted by Sticky Click here to enlarge
        Based on...? The efficiency of a twin screw over a roots has been shown at high rpm or with high volumetric efficiency whereas a roots tends to have better efficiency at lower boost from every comparison I have seen.
        There is no buildup of boost in the set example. Basically it's just the moving of the parts (and there's about the same amount in the same configuration).
        So there's no 'internal compression of air' of some sort like what Jonathan@Aviva was making believe (or at least that how I interpreted his post) and that was my point.
      1. GuidoK's Avatar
        GuidoK -
        Click here to enlarge Originally Posted by Sticky Click here to enlarge
        Based on...?
        Also if you look at performance maps of a lysholm 1600ax vs an eaton m62 (they're the most similar in the lineup in application) the low efficiency regions (you know, pressure ratio's of about 1.1 or 1.2) are about the same. (30 or 40% or so). But still that is not relevant because that is way way more boost than a bypassed compressor. At that point the efficiency would be more like 0% Click here to enlarge
        You absolutely cannot view/read the efficiency or dissipated energy from a compression map for a bypassed compressor, as it would not register (by definition that is)
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by GuidoK Click here to enlarge
        Also if you look at performance maps of a lysholm 1600ax vs an eaton m62 (they're the most similar in the lineup in application) the low efficiency regions (you know, pressure ratio's of about 1.1 or 1.2) are about the same. (30 or 40% or so). But still that is not relevant because that is way way more boost than a bypassed compressor. At that point the efficiency would be more like 0% Click here to enlarge
        You absolutely cannot view/read the efficiency or dissipated energy from a compression map for a bypassed compressor, as it would not register (by definition that is)
        I think we may be discussing two different things.
      1. Sticky's Avatar
        Sticky -
        Now that we have some real tech discussion going @LostMarine may participate again as I think it would help him to read it.
      1. DFM's Avatar
        DFM -
        Click here to enlarge Originally Posted by Sticky Click here to enlarge
        And hence leading to difficulty for the tune.. and the stock rods of course. So this setup is definitely quite different and requires different approaches/tuning.

        Exactly.