Ok I've been seeing alot of misinformation and confusion about what the Missing Link does so I thought I'd start this thread to dispell misinfo or answer any questions.

In developing the Missing Link I had many options on the table, bleed off, voltage clamping, etc. Here are the main reasons why I came to my final conclusion

Why not Voltage clamping?
1) Most map sensors are not designed to take positive manifold pressure, you will eventually pop the membrane that the strain gauge that makes up the map sensor leading you to wonder wtf is wrong with my car, because a "sorta broken" map sensor is hard to trouble shoot. voltage clamping does nothing to protect your map sensor.
2) The voltage sweep, or hysterisis leading up to clamped voltage value may not accurately reflect true real-time manifold pressure as Honda has defined it to reference in the ECU. Yes I took a Datalogger and logged values at 20ms intervals to come to this conclusion. trust me it would be a hell of a lot cheaper for me to make an electronic solution and then sell it for higher profit margin, but it doesn't work as elegantly. So with a voltclamp, your ecu is seeing one value, and producing fuel based on that value but it is not actually what is going on in the manifold.
3) Statements like this:

"there were some problems because the clamping voltage has to be adjustable for different atmospheric pressures (elevations). the J&S limiter is adjustable with a variable resistor and a led to see when the MAP voltage is below the clamp or not."

"It will have two outputs, one "limited" output and one "full" output (so you can do the afc hack or something)"

Why do you even have to worry about making adjustments, that's not the point of going turbo, the point is going fast reliably without worrying about a million adjustments except your boost controller. You should have something to put on, rely on, and go boost.

4) I get e-mails like this all the time:

"I recently installed your missing link on a g5 prelude, and it seems to work flawlessly, even better then the electronic method I use on my turbo neon; which is somewhat inconsistent on the voltage it clamps at"

Trust me, I would make more profit, and have less hassles, having designed a voltage clamp/limit solution, I would have just had to go to radio shack and get some parts and put something together, but it does not work better.

Why not regular check valves? Isn't it just another check valve?

"Correct me if I'm wrong, but isnt a missing link jsut some device that some company made so the computer wont read boost? If thats the case then why dont you just use a couple of whisper2000 check valves from your local pet store. Why spend $65 dollars when you can get the same thing with 20 dollars at most. ($2.50 per check valve at a pet store) and a dollar something for a couple of T's."

If it was that easy I would be a sheister. The valve in the link is spec'd to application, the cracking pressure, delta p, and rebound rates are matched to perform in synch with the particular MAP sensor. this wasn't something concoted out of a drunken stuppor, or drive to generate revenue. I've been making versions of the link since 1995 when I first got into the industry, I just decided to productize something that would provide a solution to everyone, so that everyone could start producing reliable power with their turbos and not have to fiddle with bullshit fixes, band-aids and all that other crap you guys spend hundreds of dollars on thinking that they will fix your problems.


Doesn't the missing link just bleed off boost like any check valve?

Wrong, the link doesn't bleed off boost by any means. The rebound rate and cracking pressure of the valve are designed to mirror the standard deviations that the map sensor normally sees. try me, take a datalogger and log the pressure values you see, with a missing link and with just fish tank or other check valves. I know that everyone here is used to jerry rigged productst that some kid in his garage came up with to capitalize on the import scene, and then called it gospel to make money. This is a purpose-built application-specific product that was designed to provide a solution to a problem.


"well its all about the install on the check valves. What i do is:
I tee off the vac line before the map sensor and put 3 check valves in. then I zip tie those check valves in. (so they dont fall out). Call me ghetto, but I rather waste my money on other things. "

Then explain to me what you do when you want to raise or lower your boost pressure? Add valves and take away valves as you adjust your boost with your boost controller, pull over, pop your hood, get under it, right before you run to add a valve because you need to bleed off more boost.

If there is one major product that the missing link works with best, it is a boost controller, there is no need for adjustments when ratcheting boost up or down on the fly. Call it a god send, but you can actually adjust boost on the fly without having to worry about your check engine light going off in the middle of a pass with a different boost setting, or on a day with a different atmospheric baro pressure, or weather or elevation change.


"i hate fish tank valves... they worked sorta.. got the missing link installed and the car runs great.. the vacume line with the CV's came off once or twice... and it sucks...
fixing car on a date sucks period... missing link 65$... no worries, happy date.. jake gets laid.... all is good..."

The bottom line . . . Thanks for the kudos.

Peter Medina
Synapse Engineering

 

I know about the missing link being a great device, there is no doubt when compared with check valves (I have one in my car in the compartment). But I have had several problems regarding the Missing Link. I may have other problems that contributed to it, but you may have answers/solutions for me.

1) My missing link was stuck open, allowing boost to reach the map. I had the brand new missing link on for a week, new turbo with a filter on. WD40-ing fixed it good and quick. No problems after that in regards to sticking for the next month until I removed it.

2) At idle and off boost, the car was running a bit rich. Using the VAFC, idle pressure with the map direct to the TB was around -550mm/Hg. After the missing link install, the VAFC was reading @ -500mm/Hg whilst the manifold was at the same pressure (20inHg vac). I plugged the bleeder hole and the pessure went back to normal.

The latter was barely noticeable, but the 1st problem was a bit of a question. Once again, the missing link is great. I just wanted to share my 2 problems I encountered, maybe it may help other users of the missing link....

 

Only thing wrong with the one I have is the vac port to the map sensor needs to be enlarged, beiung that the map sensor is quite a bit larger than the one on the missing link.

 

1) There is a "break-in" period sometimes with these valves and although it appeared to go past that period in testing, you may have experienced sticktion. Occasionally, debris, even coke, oil or carbon will keep the valve from seating properly. You are right in running some wd-40, although sometimes all it takes is to push the valve up and down with a pin or something. You may also try to clear the vent-hole of carbon with a pin. Whenever you encounter problems with the Missing Link, it is just advisable to pull it out and make sure that the valve is clear of debris, and to blow through it, somewhat the same type of care that a PVC valve needs. Why did you remove it?

2) This is part of the design, although I tried to avoid all flaws, a bit of high vaccum and idle richness was one that I couldn't avoid. Which I feel more than made-up for all of the other potential flaws. Just know that this is part of the design intent and one that cannot be avoided, however, when turbocharged, the effects are negligent to the vehicle.

Thanks for the input, just so that I know what generation of homologation your link is, tell me this, is the o-ring seat a groove, or a counter-bore? Did your link come with black and red-plugs in packaging? Thanks.

Peter

 

I hope to god that I'm reading this wrong, but if your signature is correct you do have a 91 Integ right?

If this is the case I hope that you are not trying to use an ML001 (diamond shaped) Missing Link with your car. Because your model vehicle is intended to use the ML002 unit. Take a look on our site for the installtion and configuration differences http://www.synapseengineering.com. There is no way that the ML001 can work with the 91 Integ, the ML002 must be used.

Sorry if this is way off, but I have had people try to modify the Missing Link to work with Saturns, Neons, Eclipses, etc. and not for their designed purposes.

If this isn't the case please explain because I am a bit lost with what you are asking.

Peter

 

Just to add:

1) Max boost the Missing link can take is 60 psi
2) boost can infinitely vary between 0 to 60 psi

 

Nope it is the model number 2...sorry I should have included that.

 

I guess you are talking about the brass barb fitting. Thanks for the input. It hasn't posed a problem before, but what I do recommend using for tubing material is smooth silicone that will stretch, like 4mm silicone. But not the nylon material, or the thick silicone. The thin silicone works well.

Peter

 

I removed it in favor of a home-built map limiter. I don't plan on going past 10psi anyway, so it worked out fine. I did encounter the variable-less nature of a preset voltage clamp, so I am using another oem Honda map sensor as the reference and output during boost/clamping (especially since I go up into the mountains often). For mass production, using an oem sensor as a reference is expensive, but me being cheap and the DIYer, I had to try it out. I usually reccomend the missing link for other peoples' cars, I just needed to read boost with the stock map for the VAFC mod to work out. As for it being in my compartment, it is the backup plan if all goes awry with the electronics (so far so good, though).

My version is the grooved O-ring seat. No events after the wd40-ing, it was perfect. As for the design of the bleed hole, I would have done it exactly the same, it is what makes the missing link perfect at giving atm pressures while in boost..... I wondered if the hole were even smaller if it would work, but I felt the response while going atm would lag. Added to that, I can't find a drill bit any smaller either, hehehe. Great product though.

 

i also have the missing link. the o rings are fine but my engine light always stay on. when i run the codes it throws a code 3. saying map. is my map sensor bad or is it my missing link. i cant figure it out. my car also back fires once in a while and it lags alot. my set up is b18b with a drag III.

 

Try the cleaning and WD40-ing. I used a little alcohol then wd40'd it, but ask the synapse fellow which cleaners to use. Alcohol isn't too friendly with plastics, or metal, or cuts on your hand.....

 

What year is the ECU for your b18B and in which year chassis? As long as your car is not going into limp mode the missing link is working fine. If you are using anything connected to the map sensor sensor signal to the ecu, like an AFC or anything else that is modifying the map signal, that may be the culprit. But here's the real way to tell:

1) With the missing link on disconnect anything else that affects the map and reset the ecu. if the CEL isn't coming on it's not the ML

2) With the electronics on try running the car with and without the ML, if it goes into limp mode without the ML but not with, it most likely isn't the ML.

3) Make sure that the vent hole on the side is clear of any carbon or debris, use a pin to clear it, reset ecu

4) either WD40 or alcohol will work, but if you just take the ML off and use something like an air duster, like the one's used to clean your computer to clean out the valve that will work. but all materials are resilient to alcohol, prefer wd since it may actually lubricate in the long haul.

 

how were you limiting the MAP sensor voltage? if the limiter is altering the voltage below atmospheric, then thats big news to me. please share what device you were testing. contact me privately if you don't want to name companies.

you're making some pretty serious claims here about electronic MAP limiters. are you saying your product does something different than the above statement? i thought, by definition it prevented the MAP sensor from experiencing anything above atmospheric pressure.

Joe

 

It wasn't any company's product, we were toying around with a prototype that a colleague had designed. No it wasn't clamping below atmospheric. What we found was that right as the voltage came closer to the clamping voltage that the voltage generated past the clamp was not necessarily identical to the voltage generated for that particular manifold pressure without the clamp. But again, this wasn't a finished product ready to go to market, but one we were testing for ourselves.

This is what I mean by the ecu seeing voltage past the clamp that isn't representative without the clamp. Our product also keeps the ecu from seeing anything above atmospheric, but what we found in our testing is that the pneumatic solution more accurately represented the pressure change from vaccum to atmospheric. Our product also accurately represents to the MAP sensor/ECU the true atmospheric pressure. This isn't to say that someone else's voltage clamp isn't more refined, ours was in pretty early stages, but the benefits of going pneumatic, for us, outweighed the benefits of going voltclamp.

Sorry, the second quote is a bit of misexpression,what I meant was that the value after the voltclamp going to the ecu may not be representative of the atmospheric pressure as opposed to what I said, manifold.

That is, that at high elevations, the volt clamp method will generate a condition wherein, effectively the voltage that corresponds with 0", or sea level atmospheric pressure, does not accurately convey to the ECU atmospheric pressure that is lower than atmospheric at sea level. So under boost, in high elevation, the voltage is clamping at what corresponds to be sea level atmospheric pressure, and effectively generates mixture and timing settings accordingly, when in fact there is a delta between what the ecu sees as pressure and what is actually happening at atmoshpere. This is because voltclamps will always clamp at voltage values and not atmospheric pressure.

I hope that this is clearer, but the bottom-line is that the problem with voltclamps isn't so much in the transition, which probably comes in last in order of priority to the other issues, but in changes in atmospheric pressure from what is seen at sea level or calibration. Basically you have to have a voltage to clamp to at each atmoshperic pressure below sea level. It would be interesting to take voltclamps designed and calibrated by different companies/individuals from different elevations to see what the range of voltages are that they clamp to, then you might find one calibrated at high elevation that clamps below atmoshperic when at sea level. I guess the best voltclamp would be one that could infinitely change the clamping voltage based on changes in atmospheric pressure, but then again, in the case of engines that don't have map sensors designed to operate much above atmospheric, this "ideal" voltclamp still can't keep boost pressure from popping the strain gauge's film mount.

Changes in atmospheric pressure and how that is represented to the ecu may not be a big deal to most people, but follow me through this logic. MAP sensor always tells ECU that it is sea level, then runs fuel and timing as though it is sea level, well guess what at high altitude, less air to burn, so the mixture needs to be leaner and timing more aggressively advanced, but the ECU thinks its sea level at WOT, so it is in a part of the fuel/timing map that thinks it is sea level, and the thing is running less efficient. To some no big deal, to others consistency in performance, and that little bit of the right combo of mixture and timing help for throttle response and power for the existing conditions.

For example: I think it is idiotic for a "black box" of some turbo kits to change the resistance values of the coolant temp sensor to the ECU simply to richen the mixture. But unwittingly this also gets the ECU to think that the car is running cold because the coolant is cold, but the intake air temp is normalized and stable, so what does the ECU do, advance the timing to get the car to warm up more, not so great for a turbo car with 92 octane and a warm day, love hearing those civics ping their guts outs.

I'm sure there's typos and misexpressions here apologies ahead of time, not really taking the time to carefully edit this novel.

 

Just to add and clarify, the timing and fuel maps are less affected by any of the aforementioned voltclamped conditions when in closed loop part throttle operation, but is more critical when under open-loop WOT when the MAP and tps play a greater role. In any case, this problem with the voltclamp method is further compounded by the fact that the ECU resets the reference atmospheric pressure at WOT, so that if the car is moving to many different elevations in one day the fuel and timing can adjust accordingly. Well at WOT, with a voltclamp, the ECU is always seeing the same voltage so it will never reset to compensate for changes in elevation/atmospheric pressure.

What is the bottom line, inconsistency in performance. Those of you that live in high altitudes such as Phoenix and Denver especially need to look out because products calibrated to sea level may be losing you horsepower, or look out for the converse, products calibrated at high altitude may be causing sea level users to run lean and with too much timing advance, all conditions prime for detonation.