Miata Turbo Forum - Boost cars, acquire cats.

Miata Turbo Forum - Boost cars, acquire cats. (https://www.miataturbo.net/)
-   Build Threads (https://www.miataturbo.net/build-threads-57/)
-   -   Rapid spool turbo with on-board compressed air in exhaust (https://www.miataturbo.net/build-threads-57/rapid-spool-turbo-board-compressed-air-exhaust-98666/)

Newaza 09-21-2021 06:04 AM

Quick update: I went back to track weekend before last and made a couple passes. Below was the quickest of the passes that day. It was once again a scorcher of a day so wasn't expecting to break any records. However the tune on new ecu setup is getting a little better than last outing as I'm starting to get times a little closer to what I was getting before. The performance is still not quit a match though so the tune still needs some more work, but I'm slowly dialing it in. I am getting antzy to make some of the changes I was planning before ecu swap to try and improve performance, but don't want to make any of those changes until I get the same performance with the new ecu as I did with the prior ecu with all else being equal. Once I get a match in performance I will start adding more boost, testing different gears and maybe even shooting a little nitrous.

https://cimg1.ibsrv.net/gimg/www.mia...af5298d8cc.jpg

Newaza 10-02-2021 10:37 PM

Hello Fellows.. Went to track tonight to do some testing on latest tune on the speedyefi ecu. Got a new pb on the elapsed time! Pretty stoked on that at least. Showing the tune is starting to come around. Trap speed is still off, but only slightly now, so very close. I did have a few missfires at top end of track. Might be time for new plugs, maybe even tighten the gap slightly. Could be i'm asking too much from the stock coils at this point. lol...

https://cimg9.ibsrv.net/gimg/www.mia...7b24f85b54.jpg


That converts to from 10.3 to 10.35 1/4 mile by the way.

technicalninja 10-02-2021 11:22 PM

You're within striking distance of a single digit 1/4 mile times!!!
Doing it inexpensively as well...
That's just BAD ASSED!

I've been thinking about a much smaller version of your spooler...
Anytime TPS reads WOT and boost is below a "threshold" point you get a smaller version of what you have to reduce "lag" and accelerate a normal sized turbo quicker.
A small shot should work OK with a manual.
Might be possible using a high pressure air tank alone for testing this.
It would be helpful to reach full boost 1k rpm sooner
Food for thought...

Thanks for your ingenuity and sharing it with us.

Newaza 10-03-2021 08:49 AM


Originally Posted by technicalninja (Post 1609968)
You're within striking distance of a single digit 1/4 mile times!!!
Doing it inexpensively as well...
That's just BAD ASSED!

I've been thinking about a much smaller version of your spooler...
Anytime TPS reads WOT and boost is below a "threshold" point you get a smaller version of what you have to reduce "lag" and accelerate a normal sized turbo quicker.
A small shot should work OK with a manual.
Might be possible using a high pressure air tank alone for testing this.
It would be helpful to reach full boost 1k rpm sooner
Food for thought...

Thanks for your ingenuity and sharing it with us.

Thanks for the input.

Yeah I'm thinking very high 9s may be doable. That would require lower 6.30s in the 1/8th. Trying to sneak up on that with what I have anyways. Really hoping to see Pat get his back together soon with his new setup. With his driveline and t-brake he should have no problem getting 6.30s or even quicker.

Doing all this cheaply is one requirement I have self imposed due to being married and with other responsibilities. I'm REALLY struggling with the urge to redo the entire car with a turbo smallblock setup though tbh. Been going back and forth with that in my mind for some time. Trying to maintain some self control though and not going down that rabbit hole.

One thought on doing a spool shot with a manual. When I first got this setup together and was testing one thing I noticed was if I had the converter locked up and activated the spool shot at lower rpms it would of course surge the compressor, but also if the rpm's did not climb high enough while the compressed air was being used, by the time the tank was used up the boost would fall back down again to where it would naturally be at that rpm. With the converter unlocked it doesnt seem to matter at what speed or rpm I activate the spool shot. In other words if I activate the spool shot at low speed/rpm as the boost rises, the converter allows the rpm's to rise as well, independent of vehicle speed. So when air in tank is mostly used the rpms are high enough to maintain the final level of boost. If any of this makes sense, hope I am explaining well. But I'm using a fairly large turbo for this engine size and rpm range I run at, so a significantly smaller turbo may behave better with a manual. I havent tried that yet, but I'm sure if anyone can make it work you would have a good chance of pulling it off.

Newaza 10-06-2021 07:46 AM


Originally Posted by technicalninja (Post 1609968)
You're within striking distance of a single digit 1/4 mile times!!!
Doing it inexpensively as well...
That's just BAD ASSED!

I've been thinking about a much smaller version of your spooler...
Anytime TPS reads WOT and boost is below a "threshold" point you get a smaller version of what you have to reduce "lag" and accelerate a normal sized turbo quicker.
A small shot should work OK with a manual.
Might be possible using a high pressure air tank alone for testing this.
It would be helpful to reach full boost 1k rpm sooner
Food for thought...

Thanks for your ingenuity and sharing it with us.

Hey Ninja. Since you are contemplating this I thought I would share with you a refinement I was considering. Doing as described in this rough scribbled draft may (or may not lol) work better than what I'm doing currently. If pressure is high and nozzle is shaped, sized and aimed correctly it may yield equivalent or potentially better results while using a lower volume of air.

Please excuse the scribbling. Its early and I've only one cup of coffee so far.....

https://cimg1.ibsrv.net/gimg/www.mia...2c9b7b1ace.jpg

technicalninja 10-06-2021 10:37 AM

I agree fully, having the blast hit the turbine wheel directly appears to be the best use of the "energy" in the blast.

Points against this are:
I try to NOT drill holes in the turbine housing, porting it and reshaping it are fine in my book but drilling a hole, especially a large hole, might increase tendency to crack or create a possible failure point.
Another thing that I'd worry about is "thermal shock" to the turbine wheel. After a couple of runs that puppy will be HOT! Hit it with a blast of air 1000 degrees cooler might jack with the stability of the part. You are doing this already and seem to not have issues but you are "snorting" your medicine right now and this will be akin to injecting it directly into a vein...
The super cooling of the fitting you end up screwing into the turbine housing will likely be an issue as well. This part will tend to loosen as it goes through significant temperature variations. You could weld it in but I've yet to see a welded turbine housing last any time at all.
Does your current set up require regular re-tightening of the entrance nozzle? I would expect that.

Now, I don't think I'd do this to a $1500 EFR but doing this to an inexpensive China charger would be OK. (I'd want another one on the shelf, ready to go).

Ninja rule... If you have a spare part ready to install the currently installed part will usually last far far longer. Murphy WANTS to prove you wrong.

If I add a small version of your system to my set up I'll be conservative and plumb it as you have your system plumbed now...

If I was in your shoes and had another cheap turbo ready to go I'd probably try this...



Newaza 10-06-2021 12:07 PM

Yep I have several chinachargers, and I dont think I would butcher a 1500 turbo either to test. However if it proves to be worthwhile after testing on a cheap turbo first I would likely consider it on a more expensive turbo. Especially if it turned out to be a major improvement over prior art.

I too was concerned with thermal shock somewhat when I first put this together several years ago. So far its turned out to be a non issue for my setup, and I blast mine every time I drive it. However as you imply, my current setup is a greater distance away, so that (shooting into a vein) may be an issue or may not.. Testing with a cheap turbo would be the way to go for sure and is something I may do in the future.

I have yet to need to go back and re-tighten any of the fittings going into the manifold. So far at least. Another concern I had initially was just doing a grease conversion on the modded ac compressor. I did not expect it to last. However so far its doing well and shows no signs of complaining. I just shoot some grease in it and spray some lube in the inlet occasionally.

Honestly the way I have it set up now is pretty much the way I built it initially just to test. The original plan was to quickly cobble something together to test and to re-do in a more permanent fashion if all went well. It worked so well on the first try that I never bothered to re-engineer. So it's still cobbled together crap, but at least it matches the rest of the car LOL...

technicalninja 10-06-2021 01:50 PM

You make me look at problems from a different perspective. This is a rare and good trait IMO.
Sometimes a different person's take on a problem increases knowledge far more than "standard" research.
Thanks!

Now, thinking about adding a jet to a turbine housing does bring up some application questions...
Drilling a hole at an angle to a flat surface is a difficult process. I can do it but would really prefer a real mill for this process.
Drilling a hole at an angle into something spherical is a whole nother bag of worms. I don't think I could do it without the afore mentioned mill.
Pat might be able too...
Pat would just manufacture a proper jig and do it with the mill that he built for himself. I haven't built my own mill and don't think I will ever do this (I'd just call Pat and get his take on the problem).
I expect drilling the hole may be problematic...

Next possible issue is housing thickness at the best point for the nozzle. I'd want it pretty thick at the point I did my modifications. Most turbine housings are thinner than you might expect.

And then you have routing the air supply. If you're lucky and the best place is somewhere on the upper section of the housing, you're golden.
Seldom does it work out this way and if it's on the bottom (as in your drawing) running the air supply becomes yet another nightmare...

Time to get back to fixin AC crap. I have been 12-16 hours per day, 7 days a week, since the beginning of June. I took Memorial day off. That has been the ONLY day off I've had in 14+ weeks.
I'm so freaking tired of wrenching on stuff that I can't see straight. I have eight "patients" right now...
2-3 weeks off with NO cars at all would be wonderful...

Your use of the AC compressor and the mods to get it there are highly genius.
One thing I might do is use a cotton gauze air filter for the intake and over oil that puppy regularly. The original compressor is NOT oiled by the sump. The oil that protects the pistons from the case runs around the system with the refrigerant.
You have lost this factor of a "normal" ac system. So far it doesn't seemed to have mattered but adding a small amount of oil to the incoming air charge will extend compressor life

Newaza 10-06-2021 05:50 PM

Thats a good idea on the oil soaked gauze. I think I may incorporate something like that. May even do that with a small container for light oil mounted a little lower than pump inlet, but also with an encapsulated wick leading to gauze to keep filter oiled. Might just work. thanks Ninja!

technicalninja 10-06-2021 08:21 PM

Best oil to use for improving compressor life is the old time mineral oil that we uses to use for R12 systems.
This stuff, which may be hard to find now, is very similar to mineral oil for infants so you could end up with a bottle of Johnson & Johnson baby oil in your console.
Whenever someone asks "how is your car so fast?" you can hold up the bottle and say "magic baby sauce" which will confuse them all the more...

Whatever you use (pretty much any light oil will work) DO NOT use any oil for a R134 system. It's crappy oil anyways, it is very hydroscopic (absorbs humidity), and it's fairly poisonous for humans.
Because of the hydroscopic tendencies it's easily absorbed through your skin; your liver can process PAG but your kidneys cannot and it will stay in your kidneys for the rest of your life which may be shorter because of it.
Brake fluid has similar tendencies.

Baby oil is stupid safe.
And the looks you will get will be priceless!

Newaza 10-24-2021 06:59 PM

Starting to work on spool shot version 2.0!
It will be to test what is shown in the rough drawing a few post up. I also think I am going to get another solenoid rated for higher pressure. I think there will be a benefit going from the 150psi I am using now to maybe closer to 200psi. My current solenoid is only rated for 145psi so I am slightly exceeding it, but I found one rated for 230psi for a reasonable cost.
Hopefully this will be enough of an improvement to get the gtx3576 .82 ar churbo to spool as, or hopefully even more quickly than the gtx 3076 .82 ar churbo. The 3076 has proven slightly quicker at the track while the 3576 is not quit as quick, but traps slighlty higher. That indicates that all else being equal the 3576 makes a little more power, but the lower 60 feet times compared to the 3076 shows it a little laggier, even with the spool shot.

Please don't flame for the welding.. Its an ebay housing so no biggie. I figure there is a 50% chance it will be an improvement over what I have now. 25% chance it will fall off, 90% chance I ruined this housing and 100% chance it looks like chicken crap!
https://cimg2.ibsrv.net/gimg/www.mia...97c283978f.jpg

After spool shot 2.0 is finished and tested I have several other planned changes that will happen pretty quickly. Looking to pick up a few tenths before the year is out hopefully.

technicalninja 10-25-2021 09:17 AM

Sometimes, when things are wrong and I have a bunch of plain old air in a system I've seen nearly 500 psi out of old worn out compressors.
Somewhere around that number a check valve or blow out plug functions on the compressor and releases the pressure (and refrigerant!). This is the second and final safety on a HVAC system
All R134/R1234 systems have a high pressure switch that is supposed to disconnect the clutch or "de-stroke" the compressor above 450 psi. This is the first safety on the system.
So I see compressor operating pressure range of 100-400 psi on the high side. This is in a closed system with a low side of 10-100 psi so it looks like 300psi+ gain across the compressor.

200psi is completely normal for a HVAC system to run, 300psi is not uncommon or a 100+ degree day.

But, I've never attempted high pressures on an open system. I'm interested in what your system's top pressure CAN be....

You need to be super careful with your pressure tank, solenoids, and lines at these pressures....

What can your system produce as a max pressure?

PS as an owner of a 12 unit chicken coop it's Chicken SHIT, not crap...

Newaza 10-25-2021 11:32 AM


Originally Posted by technicalninja (Post 1611395)
Sometimes, when things are wrong and I have a bunch of plain old air in a system I've seen nearly 500 psi out of old worn out compressors.
Somewhere around that number a check valve or blow out plug functions on the compressor and releases the pressure (and refrigerant!). This is the second and final safety on a HVAC system
All R134/R1234 systems have a high pressure switch that is supposed to disconnect the clutch or "de-stroke" the compressor above 450 psi. This is the first safety on the system.
So I see compressor operating pressure range of 100-400 psi on the high side. This is in a closed system with a low side of 10-100 psi so it looks like 300psi+ gain across the compressor.

200psi is completely normal for a HVAC system to run, 300psi is not uncommon or a 100+ degree day.

But, I've never attempted high pressures on an open system. I'm interested in what your system's top pressure CAN be....

You need to be super careful with your pressure tank, solenoids, and lines at these pressures....

What can your system produce as a max pressure?

PS as an owner of a 12 unit chicken coop it's Chicken SHIT, not crap...

I'm using steel tanks normally used for compressed air so should be easily safe at 200 Psi. I limited to 150psi originally because my solenoid valve was only rated to 145 psi. I know it will go over 200psi. I had the pressure cut off switch fail before and the compressor kept running when I charged the system and didnt notice it wasnt cutting off. The pressure relief valve opened which brought it to my attention. The gauge is 200 psi and it was maxed out. I cant remember what psi my pressure relief valve is, likley ~220psi., but I did build it with the safety valve originally as a forethought of what could happen. On the second drawing of the very first post on this thread it shows the layout of the pressure switch and safety relief valve. Those two rough drawings were the initial layout of the system and pretty much the way it still is.

208fabrication 10-28-2021 04:35 PM

Newaza what is your thoughts on using this type of setup with a manual transmission?

Newaza 10-28-2021 07:27 PM


Originally Posted by 208fabrication (Post 1611656)
Newaza what is your thoughts on using this type of setup with a manual transmission?

I suppose that would depend on many factors such as size of turbo and rpm in which you activated spool shot. Also pressure and flow of your "shot". I do know when I tested initially with the converter locked up to simulate a manual transmission if I activated at too low rpm it would just surge the compressor. If the rpms were high enough when the spool shot activated it seemed to work well from what I can remember. For example, if you have a turbo setup that makes 15 psi at 4000 rpm and full throttle naturally, and were cruising at 4000 rpm with very little throttle and no boost, all the sudden punched it, with the spool shot you would reach full boost much faster than without in that scenario. However if you normally make 1 psi at best at 2000 rpm naturally and 15psi at 4000 rpm naturally, and activated the spool shot at 2000 rpm you would get something more than 1 psi (not 15psi) but not really accelerating that hard as you will be in compressor surge most likely. You will not get the boost level at 2000 rpm what it normally takes 4000 rpm to reach.
It really shines with an automatic though. With a torque converter as boost climbs so do rpms, even if the car is sitting still. With my turbo size, and an automatic with no trans brake or high stall converter my car would be a dog without this system. This setup is a major player to allow me to run the numbers I do with stock rear and axles on a low very budget.

Newaza 10-30-2021 07:14 PM

Update:
I cobbled together a quick test shooting air into turbine as detailed in the rough drawing several post up. It didn't work as well as what I was doing prior, so I plug the hole in the turbine with a pipe plug and reverted back to as before. I used 3/8" soft copper, which I think was too small when factoring in the 90 degree fittings. I may revisit this idea later, but the plan now is to add more pressure and volume to test next. Below is a picture of the tubing that was tested and a picture after removing the copper tubing and re-connecting as before.

https://cimg4.ibsrv.net/gimg/www.mia...57ed41e72c.jpg

https://cimg1.ibsrv.net/gimg/www.mia...3ce087628d.jpg


The turbo that was used for this test was a gtx3576 .82 ar churbo clone. The 3576 would usually trap a little higher but would not et quit as well as the gtx3076 .82 ar churbo clone I ran most often.

The 3576 does not spool as fast as the 3076, even with the spool shot. I am looking to pick up 2 to 3 tenths so am back to working to improve the spool shot in an attempt to get the 3576 to spool as quickly, or hopefully even quicker than the 3076. Im hopeful that more pressure and volume will do the trick.

At any rate, since the 3576 was on the car I decided just to leave it on. I went to the track today with it for some test and tune. I got the fastest trap speed to date with the car. It trapped 108 mph without locking the converter. I stopped using lock up at the track as it was found the lock up clutch in the converter could not hold the power and would disintegrate after about half a dozen or so passes. The fastest the 3076 would trap in the 1/8th is 107mph, and that was locking the converter. The best the 3076 would do without locking the converter is 106 mph in the 1/8th. The higher trap speed indicates the 3576 is making more top end power. I just have to get it to 60' like the 3076. If I can that should get my very close to my goal of 6.30s in the 1/8th (high 9 second potential 1/4). I also plan to swap from the 4;10 rear in the car now back to 4;30s. That should help the 60' time and et. I also want to try pre-turbo water injection. It seems some are reporting more compressor efficiency at high pressure ratios, so Im going to cobble together a test for that too. Im on a china turbo so if I get some erosion of the leading edge of the compressor blade over time its no biggie. If it makes an improvement thats a trade off i'm willing to take.

Below is the timeslip showing best trap speed to date.
This trap speed is easilly high enough to reach my goal of 6.30ish in the 1/8th. Car makes enough power, just need to get it to leave harder and ramp in full boost more quickly.

https://cimg2.ibsrv.net/gimg/www.mia...3adc7fcab8.jpg


As an additional data point I had the dragy going on the above pass. It shows a 1/4 et of 11.05s at 76mph, and that is after letting off the gas and coasting down shortly after the 1/8th. That demonstrates that pass would have likely been VERY low 10s in the 1/4 if ran on out!

https://cimg5.ibsrv.net/gimg/www.mia...babf118bb9.png




oreo 10-30-2021 09:26 PM

IMO, that valve you are using does not look like it flows very well.

Your valve pictured below: -the electromagnet moves the plunger (red), which moves the diaphragm (purple) which allows
the gas to flow through the valve (blue). Not a very direct or large flow channel. It does not look like the plunger moves very far.
https://cimg5.ibsrv.net/gimg/www.mia...c562707f8e.jpg

If you find this to be the bottleneck, you may be able to shorten the solenoid slug to get a bit more travel on the diaphragm, or go up to the 1" solenoid or maybe 2 3/4" units in parallel might do it.
On my vehicle, I needed to open a vacuum operated actuator quickly, and found that using 2 standard valves in parallel really sped things up (YMMV).

On a side note, do you have an accumulator under the hood?




208fabrication 10-31-2021 12:15 AM


Originally Posted by Newaza (Post 1611775)
Update:
I cobbled together a quick test shooting air into turbine as detailed in the rough drawing several post up. It didn't work as well as what I was doing prior, so I plug the hole in the turbine with a pipe plug and reverted back to as before. I used 3/8" soft copper, which I think was too small when factoring in the 90 degree fittings. I may revisit this idea later, but the plan now is to add more pressure and volume to test next. Below is a picture of the tubing that was tested and a picture after removing the copper tubing and re-connecting as before.

https://cimg4.ibsrv.net/gimg/www.mia...57ed41e72c.jpg

https://cimg1.ibsrv.net/gimg/www.mia...3ce087628d.jpg


The turbo that was used for this test was a gtx3576 .82 ar churbo clone. The 3576 would usually trap a little higher but would not et quit as well as the gtx3076 .82 ar churbo clone I ran most often.

The 3576 does not spool as fast as the 3076, even with the spool shot. I am looking to pick up 2 to 3 tenths so am back to working to improve the spool shot in an attempt to get the 3576 to spool as quickly, or hopefully even quicker than the 3076. Im hopeful that more pressure and volume will do the trick.

At any rate, since the 3576 was on the car I decided just to leave it on. I went to the track today with it for some test and tune. I got the fastest trap speed to date with the car. It trapped 108 mph without locking the converter. I stopped using lock up at the track as it was found the lock up clutch in the converter could not hold the power and would disintegrate after about half a dozen or so passes. The fastest the 3076 would trap in the 1/8th is 107mph, and that was locking the converter. The best the 3076 would do without locking the converter is 106 mph in the 1/8th. The higher trap speed indicates the 3576 is making more top end power. I just have to get it to 60' like the 3076. If I can that should get my very close to my goal of 6.30s in the 1/8th (high 9 second potential 1/4). I also plan to swap from the 4;10 rear in the car now back to 4;30s. That should help the 60' time and et. I also want to try pre-turbo water injection. It seems some are reporting more compressor efficiency at high pressure ratios, so Im going to cobble together a test for that too. Im on a china turbo so if I get some erosion of the leading edge of the compressor blade over time its no biggie. If it makes an improvement thats a trade off i'm willing to take.

Below is the timeslip showing best trap speed to date.
This trap speed is easilly high enough to reach my goal of 6.30ish in the 1/8th. Car makes enough power, just need to get it to leave harder and ramp in full boost more quickly.

https://cimg2.ibsrv.net/gimg/www.mia...3adc7fcab8.jpg


As an additional data point I had the dragy going on the above pass. It shows a 1/4 et of 11.05s at 76mph, and that is after letting off the gas and coasting down shortly after the 1/8th. That demonstrates that pass would have likely been VERY low 10s in the 1/4 if ran on out!

https://cimg5.ibsrv.net/gimg/www.mia...babf118bb9.png

This is an interesting video about water/meth injection if you are able to view it..
https://motortrendondemand.com/detai...de_title/37672

Newaza 10-31-2021 12:30 AM

Hey oreo.
I am using a 3/4" orifice valve now. I am not sure how restrictive it is compared to the rest of the piping. I use 1/2" id hose from front to back, so the orifice is larger than the tubing, especially considering the length of all the tubing.
I am using two tanks mounted in the trunk totaling over 9 gallons capacity. The system pressure drops 9 gallons worth of stored high pressure air from 150psi to about 55psi in under a second with this turbo so it must flow fairly well. However I do have a second spare 3/4" inch valve and may try a second line from tanks to the turbine for two entry points, one in the manifold and one in the turbine as pictured in the current locations. However I'm going to try a larger valve rated at 230psi and add an additional air tank for more capacity first.

When you ask about an accumulator under the hood I assume you are asking about high pressure air storage tanks. Then yes, I have 2 tanks totaling over 9 gallons capacity but mounted in the trunk, not under the hood. No room under the hood for tanks this large. Volvo gets away with a small 1 gallon tank under the hood since they are using a tiny turbo that is also sequential. I'm trying to light off a MUCH larger turbo from a fairly low rpm and as quickly as possible. I'm talking of targeting from tenths to half of a second here from 0 psi to 35psi, and eventually to 40+ psi LOL!....... I'm likely going to add an additional 2 or 3 gallon tank as well. I'm thinking 11-12 gallons and 200psi storage capacity may do the trick. It takes a surprising large volume of high pressure air to light off a larger turbo as quickly as I'm attempting.

Newaza 10-31-2021 12:47 AM

Hey 208. I couldn't view the video without signing up for a free trial.. But I've seen some of the engine masters vids. Good stuff...

I will just be using straight water pre-turbo. Mainly to hopefully to increase compressor efficiency. I already have several different misting nozzles and other parts necessary to cobble something together quickly. I'll likley drill a small drain (maybe 1/8") at the bottom of each intercooler tank for any potential water collecting as well. Dont want to hydrolock lol!!


All times are GMT -4. The time now is 07:15 AM.


© 2024 MH Sub I, LLC dba Internet Brands