PDA

View Full Version : Water level indicator


leman_opc
23-08-2012, 09:55 AM
I'm currently planning an install of hfs3 system on my Opel Astra OPC (=Astra VXR in Vauxhall range), and the best place for tank/pump we found is behind the front bumper.

With the tank installed there I'll be not able to see the water level (visually, so to say), so to have an idea when to fill the tank I need a water level sensor.

My question is - from the manual I'm unsure how the supplied water level sensor works. Does the sensor activate the failsafe immediately when the low water level is detected or there is a kind of intermediate state (water is low as indicated with a blinking yellow led on the gauge but the system still working yet with no failsafe on)?

Also, would be great to know the logics behind the water level sensor operation in hfs3 (the sensor itself appears to be a discrete output one so I would imagine hfs3 controller should have a kind of algorithm to filter out the low water level signals due to simple water movements back and forth on cornering etc.)...

leman_opc
23-08-2012, 03:56 PM
Just gave a careful read to the installation guide, it says that normally failsafe on OE solenoid should be wired as white wire to ECU and brown wire to solenoid (when the OE ECU to solenoid wire is spliced).

Just before that the manual says that brown wire is normally opened link to white, and black is normally closed. So I imagine normally (when the hfs3 is on and no failsafe is triggered) the solenoid should have its pulse from ECU, so it's white wire to ECU and black to solenoid. When the failsafe is on or the whole hfs3 is off then it's no voltage to solenoid and the boost is limited to wastegate actuator breakdown pressure.

So if the black wire is normally closed it looks like the black wire should be connected to solenoid. But the manual says brown...

Lost a bit now - would really appreciate your help Richard on this...

leman_opc
23-08-2012, 04:24 PM
One last question ))) I think I saw a recommendation that the pump in line should be on one level or below the tank out, but now I can't find where I saw this...

Is this really a prerequisite for normal operation (i.e. no air in the line) or may I simply make a draw fom the bottom of the tank and locate a pump on top of the tank?

The reason for asking is that "a pump below the tank out" requirement will result in pump being fitted behind the front bumper 10cm from ground - I will obviously make some kind or dirt/water protection there but I imagine it will still be not good. If it's possible to locate the pump on top of the tank, then I may locate the tank behind the front bumper and the pump in the engine compartment (much safer / cleaner place).

Richard L
23-08-2012, 09:05 PM
I'm currently planning an install of hfs3 system on my Opel Astra OPC (=Astra VXR in Vauxhall range), and the best place for tank/pump we found is behind the front bumper.

With the tank installed there I'll be not able to see the water level (visually, so to say), so to have an idea when to fill the tank I need a water level sensor.

My question is - from the manual I'm unsure how the supplied water level sensor works. Does the sensor activate the failsafe immediately when the low water level is detected or there is a kind of intermediate state (water is low as indicated with a blinking yellow led on the gauge but the system still working yet with no failsafe on)?

Also, would be great to know the logics behind the water level sensor operation in hfs3 (the sensor itself appears to be a discrete output one so I would imagine hfs3 controller should have a kind of algorithm to filter out the low water level signals due to simple water movements back and forth on cornering etc.)...

You are correct about the level sensor, it is not just on/off.

The system monitors the duty cycle of the sensor during the low level period. The gauge led will flash in sync with the sensor's arm but will not shut the system off until the duty is over 50% for a continuous period of 20 seconds.

This algorithm will allow early warning but more cutting off the system.

If water level is low at start up, the system will not turn on, just illuminates the "low level" led on the gauge. The system can only be reset by filling up the tank.

Richard L
23-08-2012, 09:15 PM
One last question ))) I think I saw a recommendation that the pump in line should be on one level or below the tank out, but now I can't find where I saw this...

Is this really a prerequisite for normal operation (i.e. no air in the line) or may I simply make a draw fom the bottom of the tank and locate a pump on top of the tank?

The reason for asking is that "a pump below the tank out" requirement will result in pump being fitted behind the front bumper 10cm from ground - I will obviously make some kind or dirt/water protection there but I imagine it will still be not good. If it's possible to locate the pump on top of the tank, then I may locate the tank behind the front bumper and the pump in the engine compartment (much safer / cleaner place).

The pump can normally prime itself few feet about the water level. But if you allow the water level to run so low that sucking in air is imminent, the system will not recover as quickly compared to a gravity fed setup.

If you are install the pump and relay in the engine bay area, make sure the water cannot trickle along the harness and work itself in to the relay and pump. Make a downward loop along the pump harness to avoid trickling water.

You must insulate the cable gland at the back of the pump with some silicon sealant. Plastic bag up the relay cluster is possible.

Richard L
23-08-2012, 09:34 PM
Just gave a careful read to the installation guide, it says that normally failsafe on OE solenoid should be wired as white wire to ECU and brown wire to solenoid (when the OE ECU to solenoid wire is spliced).

Just before that the manual says that brown wire is normally opened link to white, and black is normally closed. So I imagine normally (when the hfs3 is on and no failsafe is triggered) the solenoid should have its pulse from ECU, so it's white wire to ECU and black to solenoid. When the failsafe is on or the whole hfs3 is off then it's no voltage to solenoid and the boost is limited to wastegate actuator breakdown pressure.

So if the black wire is normally closed it looks like the black wire should be connected to solenoid. But the manual says brown...

Lost a bit now - would really appreciate your help Richard on this...

As soon as you turn the system on, the relay is energised. This invert the "normally opened" to "normally closed" orientations. We did this because if power is interrupted at any point, the system will automatically cut boost. This is a "power fail" failsafe.


This diagram will help you:

http://www.aquamist.co.uk/HFS4/RELAY2.gif

leman_opc
28-08-2012, 09:22 AM
Thanks for your detailed response Richard!

You mentioned that the pump may prime itself few feet above the water level, but the risk of air sucking is higher. I that a serious risk or maybe it could be somewhat cured by fitting the low level sensor a bit higher above the water draw point?

Asking because my setup options are limited to the tank installation behind the front bumper (red box in the pic below) and pump installation either on top of the tank in the engine bay (relatively dry area) or behind the front bumper on the other side (green boxes in the pic below). The latter pump position will allow almost horizontal feed from the tank draw point but there is no water / dust protection behind the bumper, I may arrange something to cover from direct water splashes but it will be still hardly safe place from the dust etc. Do you think the setup with the pump on top would be a livable solution in my case?

218

Also I would appreciate your opinion on the nozzle placement. I've already decided in favor of single-nozzle setup with nozzle location after the IC but before the throttle body, so my options are basically before the MAP/IAT sensor or after it. I read here on the forums that you had experience with WMI setup on Astra Coupe with basically the same engine as mine and in that case there was a problem with nozzle spraying on the MAP/IAT. I didn't see any pics of that setup though, so I'm unaware whether it was a case of nozzle set directly opposite the MAP/IAT sensor or it was before the sensor but the W/M mixture did not evaporate well before reaching the MAP/IAT sensor. On the other hand injecting before MAP/IAT looks still good since IAT will register lower itake temps and have the ign angle set more wisely instaed of blindly running more advance implying that the WMI is on and working.

So I would highly appreciate you advice whether you think the W/M would evaporate well and not cause MAP/IAT going bad when the nozzle is placed before the sensor (red circle in the lower part of the intake), alternative placement (after the MAP/IAT) is a red circle in the upper part of the intake. MAP/IAT sensor placement is the green circle on the pic below.

219

Thanks in advance for your advice!

Richard L
28-08-2012, 08:33 PM
As long as you don't go more than 12", air will not affect the priming (4mm ID hose). It all depends on how much vacuum the pump can generate (dry condition) to lift a given mass of water up a given height. As you know, the pump is designed to pump water and not air.

I don't think you have anything to worry about on your install setup.

As far as your concern regarding spray water/meth at the MAP/IAT sensor, I suggest you ignore them and put your nozzle as far away from the throttle as possible, preferably at the exit of the IC. This will give you best charge air cooling and even cylinder distribution.

The MAP/IAT sensor is going to going be smothered the moment you snap the throttle shut. When the fast moving air hit the throttle plate, it is not going to stop but bounce around. The inlet pipe will experience a mini gale storm 10 within.

leman_opc
29-08-2012, 09:53 AM
It's always good to have the developer knowing the kit A to Z supporting the community )))

On the 1st point - did I understand it correctly that even in case of air stack 12" height in 4mm ID tube the pump is capable (though not specifically designed for) to create sufficient vacuum when working in dry condition to ensure the acceptable priming?

Richard L
29-08-2012, 11:29 AM
Yes, it does.

This is based on the the mass of water inside a 4mm ID tube. I did a calculation some time ago, you need 4" of vacuum to lift 12" of water vertically.

leman_opc
11-11-2012, 03:36 PM
It's been a while since all my questions above but managed to get the kit installed only now... Finally decided to place the pump behind the front bumper (more dust etc. but the feed from the tank to the pump almost horizontal).

For some reason I'm not able to attach the pictures.

One question though - during the install the bracket of the 40A relay was cracked, so I wanted to replace the relay with a similar one but with more heavy duty bracket (e.g. metal and not plastic). Is it just a normal 40A automotive relay or there is something special about it?

leman_opc
11-11-2012, 03:57 PM
I was looking at the TR93F-12VDC-SC-A relay ( http://lib.chipdip.ru/156/DOC000156930.pdf ) as a replacement - seems to be pretty much the same as the one that came with the kit, correct?

Richard L
11-11-2012, 11:36 PM
looks good to me.

leman_opc
12-11-2012, 07:32 AM
Great, many thanks for prompt response!

Will post the pics in the relevant forum section when everything is done...

leman_opc
14-11-2012, 01:38 AM
Got everything wired, purged the system ok, got the progressive spray on a staying car by revving up etc. - it's really nice that the manual is so detailed, really makes the life easier during the install.

Still have some questions though:
1. My SC dimmer on the gauge is turned clockwise to the max. Position but I still get 7 bars at the maximum load. Is it OK? I used the smallest jet from the kit an medium restrictor.
2. With DHB linked on idle I see no bars and no "B" symbol on the gauge. Does this mean that the failsafe is on due to flow outside the "safe window"? Is this normal?
3. Irrespective of DHB and DFS jumpers settings and system state (on, off, failsafe triggered etc) there is no contact between white and brown or white and black wires. Due to this I get a CEL light (fault code = open BCS circuit) and constant boost off mode, I believe - not tested this on the move though. In the meanwhile temporarily linked DFS and connected BCS as it was from the factory. What am I doing wrong?

Richard L
14-11-2012, 09:46 AM
answers:

1. You should only set the SC trimmer to display 5-6 bars on the gauge. This way you can monitor over flow or cut hose. Medium restrictor is fine.

2. DHB will only affect the "gauge off" mode (yellow led reminder). Only the DFS (disable fail safe) will illuminate the "B" led on the gauge, reminding you the failsafe is disabled.

3. Can you check the order of wire on the grey harness is as follows:
http://www.aquamist.co.uk/HFS3/grey harness-11.gif

leman_opc
14-11-2012, 10:55 AM
Will check the wires order later on today (thanks!).

On p.2: from the manual I had an impression that B led indicates that no boost-cut is active. I.e. even if the failsafe is operational (DFS unlinked) I should still see "B" when the system is working normally and hence the BCS is operated by ECU with no HFS-3 interference. On the initial tests when revving the staying car to get some bars displayed on the gauge I think I saw "B" light blinking once the the bars appeared. WL was turned all the way counterclockwise - so I thought once at least one bar is displayed it falls in falsafe window and voila - failsafe mode turns off.

Do I now read it correctly that "B" light should appear only in DFS mode? Hence, with DFS unlinked I should not see the "B" light and the blinking I got means something is wrong?

Richard L
14-11-2012, 11:40 AM
The "B" led (high boost enabled) only appears solid under the following conditions:

1. "B" stays on solid when DFS is linked (gauge on)
2. System is trigger... if the flow is "inside" the WL and WH. "B" l3ed stays on solid
3. System is triggered... if the flow is "outside" the WL or WH, the "B" led will turn off after a brief blink (failsafe will activate and cut boost).

Richard L
14-11-2012, 11:40 AM
The "B" led (high boost enabled) only appears solid under the following conditions:

1. "B" stays on solid when DFS is linked (gauge on)
2. System is trigger... if the flow is "inside" the WL and WH. "B" l3ed stays on solid
3. System is triggered... if the flow is "outside" the WL or WH, the "B" led will turn off after a brief blink (failsafe will activate and cut boost soon after).

leman_opc
14-11-2012, 03:50 PM
OK, understood, so before the system is triggered (e.g. the car is on idle, IDC is below the set threshold to activate WMI) it is correct that I have no "B" symbol on the gauge.

So, combining this outcome with the diagram on the previous page of this thread: no "B" led does not necessarily mean that the failsafe / boost off is active. E.g. in the situation above (the car is on idle, system not triggered due to IDC below the threshold, no "B" led on the gauge) the BCS is still controlled by ECU (white wire linked to brown). In case then the IDC increases and the system is triggered and the flow is outside of "safe window" the "B" led will blink but then stay off and failsafe mode would be switched on (white wire switched to black), when the IDC will the go below the system activation threshold the failsafe mode would be off once again (white wire switched back to black). Correct?

Richard L
14-11-2012, 04:50 PM
OK, understood, so before the system is triggered (e.g. the car is on idle, IDC is below the set threshold to activate WMI) it is correct that I have no "B" symbol on the gauge.

This is 100% correct.


So, combining this outcome with the diagram on the previous page of this thread: no "B" led does not necessarily mean that the failsafe / boost off is active. E.g. in the situation above (the car is on idle, system not triggered due to IDC below the threshold, no "B" led on the gauge) the BCS is still controlled by ECU (white wire linked to brown). In case then the IDC increases and the system is triggered and the flow is outside of "safe window" the "B" led will blink but then stay off and failsafe mode would be switched on (white wire switched to black), when the IDC will the go below the system activation threshold the failsafe mode would be off once again (white wire switched back to black). Correct?

Just to confirm you are correct:
1. "B" led does not come no at idle or the system is not triggered.
2. During this stage, BCV is still under the control of the ECU.

Also correct:
3. When the system is triggered, the "B" led will come on and stay on as long as the flow is inside the WL and WH. Brown and white still connected.
4. If the flow is outside the WL and WH, the "B" led willgo out and the white and black wire will be connected.

This diagram may help you:

http://www.aquamist.co.uk/HFS4/RELAY2.gif

leman_opc
14-11-2012, 05:07 PM
OK, great, thanks a lot for the detailed explanation.

So, my setup then works absolutely OK except for no link between white and brown or white and black wires in any state. I will have a look at the wires order in the grey harness (as advised earlier) later on today when I have access to the car.

leman_opc
14-11-2012, 08:37 PM
Hopefully found the problem - yellow and white wires in the grey harness are placed vice versa. Did not have a tester with me to check if it works OK with yellow wire instead of white as per diagram above, will do tomorrow.

Richard, do you think it may have caused any damage to the controller when it was wired with yellow wire to ECU and brown to BCV?

Richard L
14-11-2012, 08:51 PM
Absolutely no damage. the yellow wire is an input, draws almost no current.

Change the yellow and white around and let me know.

Richard L
14-11-2012, 09:00 PM
can you give me the serial number of the controller, I like to check this against others on the same batch.

leman_opc
14-11-2012, 09:07 PM
Sure, will check tomorrow and post.

leman_opc
15-11-2012, 09:07 PM
SN: 100626

http://i1230.photobucket.com/albums/ee498/lemanopc/IMG_0555_edit.jpg

Richard L
15-11-2012, 09:18 PM
The wires are definitely reversed (yellow/white). The system you got was manufactured quite sometime ago. Took you a long time to get it installed... better late than never. Looking at the flow table, it is more suitable for running two jets, ~700 to 900cc/min. You may have to set the "SC" to max. if you are using one small jet.

Can you change the white and yellow around and let me know.

leman_opc
15-11-2012, 09:42 PM
Yes, I had the system laying around in the box for almost 1,5 years )))

I indeed had to dial SC almost to the max (now winded back 2 clicks or so and it was probably too much, I think one click before max setting would be spot on). Two jets would be too much for my setup as my fuel injectors are 4*470cc/min.

Will definitely let you know the results once I wire the failsafe finally...

leman_opc
18-11-2012, 03:14 PM
Rewired the kit last night - now the failsafe is working, no CEL on BCS circuit open (I get the CEL due to "boost below normal" condition occasionally but this is due to ECU mapping). Thanks, Richard!

Also had a chance to play with injection settings, turned the threshold setting down a bit (30% IDC or so) and also dialed gain a bit higher. Noted that the maximum flow is not significantly affected with a gain setting - presumably due to my small jet size. Am I correct to assume that with a small jet size turning the gain all the way up would result in higher midrange injection (relative to IDC) but have almost no effect on the top-end injection (as it is the jet size that effectively caps the water flow there)?

If the above is correct I would assume that putting the bigger restrictor in but leaving the small jet would make the a.m. effect (non-linear injection with higher flow to IDC rate in midrange) even more pronounced, correct?

leman_opc
20-11-2012, 08:21 AM
Richard, would appreciate your comment on one additional issue: sometimes when starting the car in the dark (i.e. lights turned on before the engine cranking) I get the gauge gleaming (brightness going up and down unsystematically). If I turn the lights off the gauge turns normal (steady bright), then turning the lights off results in gleaming once again. But turning the system off and on (with a button on a gauge) makes everything fine (i.e. steady dimmed light on the gauge with lights on).

The trimmer circuit is wired from my stereo dimming circuit, which is in turn based on a CAN-bus adapter. My dimmer trimming wheel on HFS-3 controller is turned almost all the way counter-clockwise. I note that the manual says to hook the dimmer wire to the headlamp and not the illumination circuits, may this be the reason? Is there any significant current in the dimmer circuit on HFS-3 so that it would make sense to rewire it to constant +12v via a relay triggered by ILL signal?

--------
EDIT: I stand corrected on above - gauge gleaming occurs when starting the car on cold (after it spent some time in the cold air and cooled down completely, +2 or +5 degrees C currently in Moscow), turning the HFS-3 on and off or turning lights on and off does not help. However, after some minutes the car gets warmer and the gleaming disappears - steady dimmed light with the lights on. The systems works otherwise perfectly. I will check if my battery is maybe low today but the car starts normally.

Richard L
22-11-2012, 02:27 PM
The H3's dimming circuit is not that hi-tech. It is activated by a steady 12V signal. Most cars controls the lighting via canbus but these signal is not suitable for the HFS3.

Wire the H3's dimmer wire directly to the headlamp makes it simple.

Thank for helping out with failsafe on the other thread.

leman_opc
22-11-2012, 04:06 PM
When I mentioned that I hooked my HFS-3 dimmer input to the CAN-bus I meant that I used an ILL out on a CAN bus adapter for Alpine stereos. This CAN bus adapter should exactly produce a +12v on its ILL out (I'm not sure whether it does, however).

Do I read it correctly that oscillating voltage on dimmer wire of HFS-3 controller may result in my dimmer issues?

PS. always happy to help - basically just conveyed the info you provided to me earlier, you should keep an eye on what I'm posting there however as I'm not much into electronics at all )))

leman_opc
04-12-2012, 08:49 PM
Richard I would really appreciate your help on the gauge flickering issue above. I made a short video just to show how it looks like: http://youtu.be/lEWS4KTjCog

I also had a strange issue today: each time the system activates and start spraying (I get just "B" appearing or maybe also one bar on the gauge) the "water level" led lights on and then the "w.injection" led lights and the failsafe activates. On no throttle then both lights disappear. If you accelerate lightly (so that the IDC is somewhere around the trigger) you see the "B" led flickering just as normal.

I noted that on the go the problem above occurs but if the car stands still for a while with the engine running it sprays OK for some time, you may get a few bars on the gauge etc. But then after a short run the problem appears once again.

It's -5 in Moscow now and I first thought it could be due to ice particles in the 4mm tube blocking the FAV turbine or something. But I have vodka in the tank, so doubt it could freeze at -5C, anyway it's not an explanation to "water level" led turning on.

Richard L
04-12-2012, 10:14 PM
Can you answer the following:
1. Where are you connecting the red wire of the molex connector. Fluctuation of the 12V supply?
2. The serial number of the unit.
3. Are you using the purple wire of the molex connector?
4. Is your circuit board exposed? Condensation?
5. Check the gauge cable is secure.
6. Open up the gauge and check the soldering of the RJ connector
7. Blockage will not cause the gauge to flicker.
8. We have not tested the system with vodka

One remote possibility:
As the water/vodka mixture depressurises at the nozzle tip, it will reduce the temperature further. This may cause the partial freezing at the orifice of the jet and reduce flow. This happens quite often in Canada at this time of the year. This problem can normally be solved by upping the concentration of the methanol ratio. In your case, putting more vodka into the tank. Let me know if it works.

leman_opc
05-12-2012, 11:20 AM
Can you answer the following:
1. Where are you connecting the red wire of the molex connector. Fluctuation of the 12V supply?
------
I have a dedicated power distributor for all aftermarket gauges as well as HFS-3 (based on a solid state relay) the feed to relay is from the battery (0.7m length AWG 4 wire). I did not test but doubt there is major fluctutaion - all the gauges on the A-pillar in the video use this feed as well and no flickering.
------
2. The serial number of the unit.
------
SN: 100626
------
3. Are you using the purple wire of the molex connector?
------
Yes, and the flickering is only when the hedlights are one (dimmed mode). When the headlights are off the gauge is lit with steady bright light (i.e. works perfectly fine).
The purple wire is connected to ILL output of the CAN-bus adapter for Alpine stereos (+12v out when lights on). I didn't check if there is any fluctuation on this output however and do not know which charge it is designed for. On the other hand I have a bunch of electronics hooked to this ILL out (incl. other gauges on the A-pillar) and it works perfectly for them.
Is there any significant current in HFS-3 dimming circuit (purple wire)?
------
4. Is your circuit board exposed? Condensation?
------
Nope, it's in the standard casing, located in the cabin (due to temp fluctuations some condensation si possible but not much I believe).
------
5. Check the gauge cable is secure.
------
Will do (but as mentined above all semms to work OK except the dimmer).
------
6. Open up the gauge and check the soldering of the RJ connector
------
Will do (as a last resort :))
------
7. Blockage will not cause the gauge to flicker.
------
Understood, 2 completely different issues.
------
8. We have not tested the system with vodka
------
Vodka is basically 40% ethanol, 60% water. I double-checked there were no additives in what I used.
I will check though if increasing methanol ratio would help (today after a night in a warm garage the system worked perfectly normal). What I fail to understand is why the "water level" indicator lights up if the whole issue is due to vodka freezing.
------

Richard L
05-12-2012, 11:43 AM
Aquamist dimmer circuit:

The circuit PWM's the gauge led, enabled by a solid 12V signal from the purple wire. The dim trimmer on board just attenuates the incoming signal. This controls the duty cycle of the PWM driver to the gauge leds.

If the incoming 12V signal is modulated, it will have unpredictable results. If this is the case, connect the purpose wire directly to the head lamp bulb.

I think the flow problem is possibly associated with semi-freezing transition at the nozzle tip. We have measured the tip temperature of the jet at 20C room temperature, we recorded a 5-degrees drop.

leman_opc
05-12-2012, 12:07 PM
Richard, many thanks, understood re. dimmer problem. Will double-check.

But why the "water level" led lighting when the w/m gets frosen on the nozzle? I would assume that I would only get no injection, no bars on the gauge and eventually "w.injection" led on when the failsafe is triggered - but no "water level" as the water level sensor should no be affected...

Richard L
05-12-2012, 12:13 PM
The water level led doubles up as a flow problem as well as low level indicator. Do you have a v12 gauge with two yellow leds?

Forgot to ask.. is vodka more readily available than methanol in Russia?

Richard L
05-12-2012, 12:34 PM
I also had a strange issue today: each time the system activates and start spraying (I get just "B" appearing or maybe also one bar on the gauge) the "water level" led lights on and then the "w.injection" led lights and the failsafe activates. On no throttle then both lights disappear. If you accelerate lightly (so that the IDC is somewhere around the trigger) you see the "B" led flickering just as normal.

I noted that on the go the problem above occurs but if the car stands still for a while with the engine running it sprays OK for some time, you may get a few bars on the gauge etc. But then after a short run the problem appears once again.


This strange phenomenon is quite common on DI engines. Once a while, the ECU injects a small amount of fuel into the engine to clear the trapped nitrous in the cat. The aquamist picks up this signal when it happens. You can put the jumper link to FDC and watch the workings of the fuel injectors.

Here is a link (http://www.aquamist.co.uk/forum/BMW-nox-trap.pdf) to make this clearer. Manufacturers use a fancy name "LNT" regeneration or we call it "fuel dumping".

leman_opc
05-12-2012, 03:39 PM
The water level led doubles up as a flow problem as well as low level indicator. Do you have a v12 gauge with two yellow leds?

Forgot to ask.. is vodka more readily available than methanol in Russia?

OK, I see, didn't know that (thought these are two separate failsafe mechanisms basicaly: once you have low water level you get "water level" led lighting and failsafe on, once you have flow problem you get "w.injection" led and failsafe on).

Yes, I have a gauge with 2 yellow leds (red leds on the flow bar). Not sure about the version though - will check whether it says anything on the back etc.

You may say vodka is more readily available as you just go to the grocery store and buy it. For methanol it's more complicated as it's applications are limited to technical purposes, so you should search for a source. The other question is that it's more expensive (the one I used was ca. GBP 4 per 0,5 litre bottle) and somewhat less effective - so in the end I'll have to go and find where to get meth ))) I thought it's the same in UK (except for probably higher vodka price)?

Richard L
05-12-2012, 05:15 PM
OK. We have a good reason for the dual function.

- water level only problem ......... one led on
- flow problems (failsafe activation) ....... both leds on.

It is easy to identify "one" or "two led" when you are watching the road. We tried a different colour for the failsafe but it can be quite confusing to identify when you have a bar graph with a different colour.

So we decided to use two colours only for the gauge. Yellow/2-yellows = problem (a standard warning signal colour).


Vodka in Russia ........ luck you! You can drink it to keep warm and make more power for your car.

Richard L
05-12-2012, 05:29 PM
Also had a chance to play with injection settings, turned the threshold setting down a bit (30% IDC or so) and also dialed gain a bit higher. Noted that the maximum flow is not significantly affected with a gain setting - presumably due to my small jet size. Am I correct to assume that with a small jet size turning the gain all the way up would result in higher midrange injection (relative to IDC) but have almost no effect on the top-end injection (as it is the jet size that effectively caps the water flow there)?

If the above is correct I would assume that putting the bigger restrictor in but leaving the small jet would make the a.m. effect (non-linear injection with higher flow to IDC rate in midrange) even more pronounced, correct?

The gain trimmer (probably badly notated on our part). We should have called it "F-idc% re-scaling trimmer". A bit too long to put into the limited board space underneath the trimmer.

For example, if your maximum injector opening is 50%. increase the gain by x2 will allow a full linear meth flow progression (100%)

But if you set gain trimmer to x4. You will reach 100% meth earlier. More throttle will only result in the same 100% flow.

http://www.aquamist.co.uk/HFS3/GAIN.gif

leman_opc
10-12-2012, 08:28 AM
OK. We have a good reason for the dual function.

- water level only problem ......... one led on
- flow problems (failsafe activation) ....... both leds on.

It is easy to identify "one" or "two led" when you are watching the road. We tried a different colour for the failsafe but it can be quite confusing to identify when you have a bar graph with a different colour.

So we decided to use two colours only for the gauge. Yellow/2-yellows = problem (a standard warning signal colour).


Vodka in Russia ........ luck you! You can drink it to keep warm and make more power for your car.

2 leds lighting consequently - clever solution! It's always a bunch of experience and testing behind simple solutions like this one...

Is it supposed to turn on the "water level" led always first and only then a "w.injection" led?

PS. Thanks for the IDC gain clarifications

PPS. Yep, just have a couple of bottles of vodka in the trunk and you are prepared to any unexpected occurrencies ))) This reminds me - I never posted the pics of my install (will do), I've also a couple of pics from my first system fill-up - caught a couple of shocked eyes on a parking lot )))

Richard L
10-12-2012, 09:13 AM
Do post some pictures please. Especially the filling up!

Howerton Engineering
12-12-2012, 06:33 AM
How much is a liter of Vodka in Russia? In the USA, we typically pay $30usd for 750ml, but very cheap brands might be had for $15 for 3.8L.

Methanol in bulk, about 30 miles away from us is $3.50 per 3.8L.

But I do think in a emergency a few extra bottles of vodka in the trunk can certainly remedy many issues, including an empty tank!

leman_opc
13-12-2012, 06:41 PM
The one I got was ca. 400 rub/liter (ca. $13). The cheapest one I saw was like 250rub/liter ($8 or so) but it had sugar listed as an ingridient so I decided in favor of more fancy brands )))

UPD. made and install thread with the pics finally http://www.aquamist.co.uk/vbulletin/showthread.php?t=2393

leman_opc
06-04-2013, 04:12 PM
Noted today that the pump is energised only once the IDC trigger is reached an injection starts. Is this how it is supposed to be?

I previously thought that the pump should be working all the time for better system response purpose (irrespective of whether any injection actually occurs)...

leman_opc
08-04-2013, 08:30 AM
BTW I changed my pump for a new one this weeked and I am pretty sure the flow issues I had previously (normal injection couple of seconds, then failsafe activation; or even no flow at all) were due to dying pump and not due to injection mixture freezing. As the pump is activated only when the injection actually starts it is highly important that it spools up fast enough to keep the pressure in the lines; this seems to be a problem when the pump motor shaft bearing starts to fail (i.e. it is not seized fully but the motor spool time is increased).

Interestingly, I had my old pump disassembled and cleared the debris from the bottom shaft bearing (it was completely seized by then) so that it could work at least somehow until I replace the bearing. The spool time was still bad - but I arranged the feed from the battery directly (i.e. the pump was on all the time, whether there was any injection or not) and this setup worked OK for me for a couple of evenings. So I'm pretty confident it's the spool time being significant enought for the failsafe to kick in which was my problem previously. Probably, if I was injecting enough for the pump capacity to become critical this would be another problem I'd face with a failing shaft bearings...

With a new pump there is no problem anymore even with the pump powered up as should (i.e. only when the injection starts).

Richard L
08-04-2013, 10:55 AM
The most probable place for the water entering the pump is at the cable gland. If you put some silicon sealant around that area, it will help a great deal.

Are you replacement the older pump with a aquamist pump or from a third party supplier. All non-aquamist pumps run at a slower speed, higher cam profile and very strong bypass valve spring. It makes the line pressure very spiky and draws very high currents. It you adjust the pressure down, the pressure spike gets bigger.

leman_opc
08-04-2013, 12:19 PM
Re. silicone sealant on a cable gland - I remember your advice on this, and I asked for this during the installation but for some reason it was not done...

This time I made sure I do this myself...

http://opc-club.ru/vb/index.php?app=core&module=attach&section=attach&attach_rel_module=post&attach_id=64725

http://opc-club.ru/vb/index.php?app=core&module=attach&section=attach&attach_rel_module=post&attach_id=64726

PS. the pump is an Aquamist item, while dealing with water injection on my car I actually gained a great confidence in Aquamist products so won't go with anything else when I had a chance to get an original replacement part...

Richard L
08-04-2013, 04:19 PM
This one should last a bit longer.

leman_opc
09-04-2013, 02:05 PM
Richard, I recently tried a twin jet combination intead of single jet (0,9mm -> 0,5mm + 0,3mm) and encountered a non-liner flow problem. Actually, the flow tends to burst for a couple of seconds but on a prolonged run under load the flow gets even lower to the upper revs.

On a single jet setup there was no such issue. Apart from jets and 4 mm piping to them nothing was changed.

The jets are connected via a tee supplied as a part of HFS-3 kit, after a tee there is one jet on a 7 cm hose, another jet is linked via ca 1m hose with a 15 psi checkvalve on it (also Aquamist part).

I'm now thinking whether having a 15psi checkvalve on one jet and no checkvalve on the other could cause the first jet to be blocked after the initial spray period. I don't see a good theory behind this - but this could explain flow decrease.

Did you ever experience similar issues? I'm currently thinking of getting a second checkvalve so that both jets are connected in a similar manner...

leman_opc
13-04-2013, 08:42 AM
Added second checkvalve and the problem is gone. I suspect initial "overinjecting" period was due to cyphoning effect and water from 4mm line flowing out via preturbo jet, hence some seconds needed to fill the empty line before actual injection starts next time the system is triggered and in these seconds FAV having much lower counter pressure from jets / checkvalves side.

Richard L
18-04-2013, 08:30 AM
you are 100% correct.