Cummins 4BT & Diesel Conversions Forums banner

Calculating 300hp using cc/ 1000 shots and turbo sizing

72K views 145 replies 16 participants last post by  chenchobowden 
#1 ·
I am starting this thread to make sure I am doing this right (as there are ppl on here that know a heck of a lot more than I do), and for those that don't want to take someone's word for how much fuel they are gonna need to produce a certain amount of power.

I am trying to nail down the cc/1000 shots that I need to turn my pump to for my desired horsepower. On the way to this, we can also find how many lb/min of fuel we will need though out the RPM range, which, with your A/F ratio, will give your lb/min of air for turbo sizing (not factoring VE).

I am going to use the 4BTA CPL tag 1963 which is pretty much the same as my 1839 tag as far as the p pump goes. These tags max out at 2500 RPM, so I will interpolate the data at higher RPM.

The BSFC (Brake Specific Fuel Consumption) is the rate of fuel consumption divided by the power produced. On the EPA tag it is given for us as:

lb/(HP*hr)
and
g/(kW*hr)

I will be using imperial for this example.

I want 300 HP out of my 4BTA @ 3200 RPM

On the EPA tag, we can follow the BSFC curve up and approximate it at .42 lb/(HP*hr), which is the fuel consumption at full load at that RPM. So now let's find the lb/hr of fuel:

300 HP * .42 lb/(HP*hr) = 126 lb/hr of fuel (divide that by 60 min and multiply your A/F ratio and you got your tentative airflow)

126/4 cylinders = 31.5 lb/hr per cylinder

31.5 lb/hr * (1 liter/1.834 lb) * (1000 cc/ 1 liter) * (1 hr/ 60 min) = 286 cc/min assuming 1 liter of diesel weighs 1.834 lbs

286 cc/min * (1 min/ 1600 shots) * (1000) = 179 cc/ 1000 shots 4 stroke engines inject fuel every other stroke 1600 = 3200/2

That is how much fuel needed to get 300 HP with those conditions. Now let's look at max torque. I am aiming for 600+ ft*lbs @ around 1800 RPM.

Formula for HP from RPM: HP = Torque*RPM/5252 = 600*1800/5252 = 205 HP

205 HP * .334 lb/(HP*hr) / 4 cylinders = 17.1 lb/hr per cyl

17.1 lb/hr * (1 liter/1.834 lb) * (1000cc/1 liter) * (1hr/ 60 min) = 155 cc/min

155cc/min * (1 min/ 900 shots) * (1000) = 172 cc/ 1000 shots

So I will need 172 cc/ 1000 shots to get 600 ft*lbs of torque at 1800 RPM.

This is pretty simple, a couple weeks back I thought there was some secret to get the cc/1000 shots, but someone smarter than I am pointed me in the right direction (i.e. the EPA tag) and this is what I got. Let me know if I am missing anything here.

-Ryan
 
See less See more
#2 ·
Alrighty then.

cc/1000 shots is about fuel delivery.
Engine efficiency (BSFC) determines how much of that fuel gets turned into usable shaft torque.
Power is torque x rpm (with correction for units used)

The EPA tags give you rpm, cc/1000 shots and power.
So you can crunch the power and rpm down to torque.
The cc/1000 shots then gives you the efficiency.
But first we must crunch some numbers. I work in metric. It's easier.

You're using 0.42lb/hp/hr. In metric that is 257g/kwh. The numbers I've previously been using at ~3,200rpm are within a few percent of that.
I'm using VE (volumetric efficiency) of 0.74 at that rpm.
300hp = 212kw
Take 257g/kwh of diesel, multiply it by 212kw and we get 54.48kg of diesel per hour or 908g/min
Using an A/F ratio of 18 then we need 54.48 x 18 = 980.7kg of air per hour or 16.34 kg/min.

Fuel

At 3,200rpm in a 4 cyl 4 stroke diesel you've got 6400 bangs per minute.
So into each cylinder we have 908/6400 = 0.142g of diesel and 2.55g of air.

I use 0.85 g/cc for diesel, so that's 0.167cc per bang or 167cc/1000 shots.
That's within a few percent of yours. So all good there.

Air

We need to fit ~2.55g of air into a cylinder of 0.95 litres volume.
Air at 20C and 101.3kPa has a density of 1.205kg/m^3 or 1.2g per litre.

With no compression we would only have 1.205x0.95 = 1.145 grams of air in the cylinder.
To fit in our 2.55 grams we need to increase the density by 2.55/1.145= 2.22 times.

But it's far worse than that. No engine can fill it's cylinders perfectly at all rpm. In this case (2 valve diesel) I think the VE can be as bad as 74%. So each stroke only gets 0.74 of the air it should.
So we actually need to increase the density by 2.55/(1.145x0.74) = 3 times.

To get 3 times as much air in, we need to compress it at least 3 times and then cool it (intercooling) as much as we can.
Using an 80% effective intercooler and a 70% efficient turbo compressor I get numbers like 33.5psi of boost required.
The air out of the turbo compressor will be around 190C. The air out of the intercooler will be around 54C. Airflow is ~36 lb/min.

Turbo
To run the turbo compressor to compress that much air to that much pressure takes ~46.2kw (62.5hp) worth of shaft power. Your intercooler is shedding 37kw of that as waste heat.
I figure your EGT at this point will be somewhere around 700C. I'm using 715C for this example.
Now a perfect sized turbine of 70% efficiency to extract just the right amount of power from the exhaust gas would be running at an expansion ratio of ~2.7:1 and a corrected mass-flow of ~24 lb/min. Now this is a little bit smaller than a HX35-12cm turbine.

So if you were to only run an engine at this power rating, you could bolt on a 12cm HX35 turbine and you'd be almost good. But of course the HX35 doesn't do much the rest of the time on a 4BT.
Turbine drive pressure would be around 31psi, just a shade below required boost.

I'm out of time tonight. I'll have a dig around for more suitable turbos tomorrow and see if I can map one out.
 
#4 · (Edited)
Alrighty then.

cc/1000 shots is about fuel delivery.
Engine efficiency (BSFC) determines how much of that fuel gets turned into usable shaft torque.
Power is torque x rpm (with correction for units used)

The EPA tags give you rpm, cc/1000 shots and power.
So you can crunch the power and rpm down to torque.
The cc/1000 shots then gives you the efficiency.
But first we must crunch some numbers. I work in metric. It's easier.

You're using 0.42lb/hp/hr. In metric that is 257g/kwh. The numbers I've previously been using at ~3,200rpm are within a few percent of that.
I'm using VE (volumetric efficiency) of 0.74 at that rpm.
300hp = 212kw
Take 257g/kwh of diesel, multiply it by 212kw and we get 54.48kg of diesel per hour or 908g/min
Using an A/F ratio of 18 then we need 54.48 x 18 = 980.7kg of air per hour or 16.34 kg/min.

Fuel

At 3,200rpm in a 4 cyl 4 stroke diesel you've got 6400 bangs per minute.
So into each cylinder we have 908/6400 = 0.142g of diesel and 2.55g of air.

I use 0.85 g/cc for diesel, so that's 0.167cc per bang or 167cc/1000 shots.
That's within a few percent of yours. So all good there.
I definitely agree. Metric is easier, but not as useful for me since I live in the States. So I was recrunching your numbers to see why we didn't end up at the same cc/1000. Turns out your fuel weighs just a little bit more than mine, but even with that correction we were still a little off. Just went through and converted all the numbers again and it looks like 300hp is 223 kW, not 212 kW. So using 223 kW instead and if we both use the same weight for fuel, we end up at the same answer of cc/1000.
 
#5 ·
Right you are, ashville! That's exactly what I will be doing. Just need to continue calculating and try to understand the flow and the PR of the LP turbo compared to the HP turbo through the RPM range. I want to maintain turbo efficiency at high RPM and at cruise if I can. This post is just one step in the process to get there. Hence, me, picking everyone's brain on their experience AND knowledge.
 
#7 ·
This is all good but what are the options for the low budget people that can't send a pump to the shop to have set at a specific cc? Not every one has down time or money to have their pump set to these settings.
 
#8 ·
You don't need to send the pump anywhere.
Once we've worked out the fuel and boost required, you can set the boost, check your IC is working well, tune in your EGT or A/F ratio and you'll be in the ballpark. Go for a dyno run if you want to do a live final tune and check.

We already know that a VE pump maxed out can deliver ~180cc/1000 shots. This is looking like plenty for 300hp.
 
#9 ·
Still not every one is running A/F gauges and everything else you guys are using. Now if you were to use total fuel plate movement (ppump) Afc pressure setting, depth of pre boost screw, and if the Afc is set all the way forwards or towards the rear of the engine. That's what everyone would be able to work with.
 
#11 ·
Fuel plate movements will vary too much from pump to pump and have no relevance to other pumps. Where cc/1000 shots is good across the board.

But with the trifecta known of cc/1000 shots, boost and EGT, you can simply set boost with EGT and you're good enough. If you don't have at least a boost and EGT gauge, then you can't really play this game.
 
#16 ·
So step 1 for tonights work is to change the power to 222kw (300hp at 740 watts per hp = 222kw).

Results are:
Fuel:
175 cc/1000 shots (getting near the limits of a VE pump).
Air
0.285kg/s (37.7 lb/min).
Boost and Turbo Compressor
36psi.
Turbo compressor PR is 3.45 (getting high for a single).
Air out is 198C
Intercooler air out is 56C.
Turbo shaft power is 50.9kw.
Intercooler heat shed is 40.8kw.
Turbine
Corrected choke flow (this number is basically the size of the turbine) 22.1 lb/min
EGT around 717C.
Turbine PR of 2.89
Drive pressure of 34.4psi (including exhaust outlet pressure of 117Kpa or 17psi giving 2.3psi exhaust restriction).

So from here I want to apply the same fuelling at 2000rpm and see what turbo is capable of that.

2000 rpm investigation:
Variables.
VE I'm using 0.87.
BSFC I'm using 214g/kwh.
Turbo compressor is 70% efficient.
Intercooler is removing 80% of the heat (80% effective).

Results are:
Torque & Power:
797Nm & 167kW.
Fuel:
175 cc/1000 shots (intentionally the same as before).
Air
0.179kg/s (23.6 lb/min).
Boost and Turbo Compressor
27.7psi. (lower due to better VE at lower rpm.
Turbo compressor PR is 2.88
Air out is 168C
Intercooler air out is 50C.
Turbo shaft power is 26.5kw.
Intercooler heat shed is 21.2kw.
Turbine
Corrected choke flow (this number is basically the size of the turbine) 17.7 lb/min
EGT around 710C.
Turbine PR of 2.38
Drive pressure of 23.6psi (including exhaust outlet pressure of 111Kpa or 16psi giving 1.3psi exhaust restriction).

So, at 2000rpm to burn all the fuel we need a smaller turbine on our turbo than the ideal size for max power at 3,200rpm.
The usual way around this is to fit the turbine that gives the required boost at 2000rpm and use a wastegate to bypass some exhaust at higher rpm. This works quite well, the one downside is the smaller turbine has to produce the same power from less flow at 3200rpm (because some flow is diverted around it via the wastegate). This gives higher drive pressures which can cost some power.

Fitting the smaller turbine
So what happens if we put a turbine of the ideal size for 2000rpm max torque on and run it at 3,200rpm?
My calculations show about 9% of the total exhaust needs to be bypassed around the turbine by the wastegate. The turine pressure ratio rises from the 2.89 earlier (giving 34.4psi drive) to 3.28 (41psi drive).
So now our drive pressure is about 5psi higher than boost. This is perfectly acceptable, but it does cost some power. It takes ~8.2kw of power (9hp) to make up for the extra flow restriction. So now our engine likely won't quite make 300hp without some extra boost and fuel. Where with the bigger (but laggier) turbo it would do 300hp.
 
#28 · (Edited)
So, at 2000rpm to burn all the fuel we need a smaller turbine on our turbo than the ideal size for max power at 3,200rpm.
The usual way around this is to fit the turbine that gives the required boost at 2000rpm and use a wastegate to bypass some exhaust at higher rpm. This works quite well, the one downside is the smaller turbine has to produce the same power from less flow at 3200rpm (because some flow is diverted around it via the wastegate). This gives higher drive pressures which can cost some power.

Fitting the smaller turbine
So what happens if we put a turbine of the ideal size for 2000rpm max torque on and run it at 3,200rpm?
My calculations show about 9% of the total exhaust needs to be bypassed around the turbine by the wastegate. The turine pressure ratio rises from the 2.89 earlier (giving 34.4psi drive) to 3.28 (41psi drive).
So now our drive pressure is about 5psi higher than boost. This is perfectly acceptable, but it does cost some power. It takes ~8.2kw of power (9hp) to make up for the extra flow restriction. So now our engine likely won't quite make 300hp without some extra boost and fuel. Where with the bigger (but laggier) turbo it would do 300hp.
Excellent! We have a single turbo that will fulfill our needs for around 300HP and 600 ft lbs of torque. From the wheel sizes and the a/r, it looks to be right in between a HX30 and HX35. I'm pretty stoked to see the results from J Mack. Lucky man.

As much as I would want to snag a BW 6258 turbo, I frankly can't afford it. Right now I have a HX30w6 with a 40mm wheel. With the points that Dougal plots on Matchbot, at point 3 @ 2000RPM I am at the upper PR limits of the HX30. After that I am over speeding my turbo. That is at sea-level, thus at altitude my compressor map shifts to the left and my PR's increase which is worse for me at higher RPMs. So I can get fairly close to my 600ish ft lbs with my turbo but I won't be able to get the power I want out of the higher RPM range. So I need something else... compounds.

Dougal, I am interested in how you calculate the turbine side of things. Is there a formula for flow on the hot side? From what I understand, PV = nRT applies to the combustion process of before and after the combustion. The T (temp) increases, R is constant (will change slightly for the different gases), P (drive pressure) we want to be the same as boost 1:1 and should be constant. If V (volume) stays constant, then [n], the flow (lb/min or kg/s), would decrease proportionally with T right?
 
#17 ·
I threw the above numbers in the Borg Warner Matchbot. It's a great way to check your work, particularly on turbine sizing:
http://www.turbos.bwauto.com//after...wts=400&pt6_wd=83&pt6_wd2=74&pt6_wrsin=92044&

Turns out 17.7lb/min is the size of the turbine on the EFR6258 with the 0.8 A/R turbine housing. What is quite interesting is 3200rpm appears past max power. Matchbot is showing slightly more power at 2800rpm. The exact point will depend largely on the engines air consumption and VE.
Port and polish work will move your max power to slightly higher rpm.
 
#20 ·
I threw the above numbers in the Borg Warner Matchbot. It's a great way to check your work, particularly on turbine sizing:
http://www.turbos.bwauto.com//after...wts=400&pt6_wd=83&pt6_wd2=74&pt6_wrsin=92044&

Turns out 17.7lb/min is the size of the turbine on the EFR6258 with the 0.8 A/R turbine housing. What is quite interesting is 3200rpm appears past max power. Matchbot is showing slightly more power at 2800rpm. The exact point will depend largely on the engines air consumption and VE.
Port and polish work will move your max power to slightly higher rpm.

I should be able to put Dyno numbers to Dougal’s theatrical numbers soon, I’m in the process of building this exact engine with a P-pump and Borg Warner EFR6258, the pump is at the shop now and they will set it for 178 cc/1000 shots.
 
#18 ·
Are you figuring in anything for elevation? I know that to run similar sized turbos someone at sea level can use larger turbines and housings where someone in Colorado in the mountains needs smaller housings and wheels. The s472 that was on my 24v ran great in Texas but when I got it to Missouri it was a Smokey pig, and that was roughly 2000ft difference in elevation
 
#27 ·
That plot is for sea-level. Sea-level is always the starting point, then you look at derating or re-evaluate turbos depending on altitude after that.

Here I one I've done previously for J-Mack at almost 3000ft altitude: http://www.turbos.bwauto.com//after...wts=400&pt6_wd=83&pt6_wd2=74&pt6_wrsin=92044&

It is down to 290hp at 3,200rpm and ~280hp at 2,800rpm. Also running a bit hotter and richer.

Altitude means the air is thinner and air-flow through your engine is lower. This reduces power/torque and also makes it harder to spool up turbos. Your turbo runs a higher PR to deliver the same boost so it can end up off the map and risk over-speed.

Compounds are almost a requirement for performance builds at altitude. It's the best way to get enough pressure ratio capacity without overspeed and also get early spool with good flow capacity.
 
#19 ·
Excellent point Ppump. I am in Utah. That's 4500ft. So yes, I will factor that in, eventually. But it is easier for people to go from standard pressure to their local pressure, than to go from my local pressure to someone else's. We're just creating a baseline for people to work from.
 
#23 ·
I know you guys are doing all this scientific stuff and I understand most of it but the vas majority of the members and people out there don't. That's why I wanted to make sure that the info of the hard parts are stated.

Like on my 12v that I had (5x12 set at 300 bar, 16° timing, non- intercooled, 62/71/.80, 131 DVs, stock lift pump set @ 40 psi, max boost was 42 psi, stock fueling, no fuel plate, Mack plug,) originally a 190 hp bus engine, put down 440 hp at the rear wheels last time had it on the dyno.
 
#25 ·
While there are lots out there that do not understand, there are many, like myself, that are interested in, and are working on learning, the science behind selecting the correct turbo and tuning for it.

That said, there was no scientific reason for installing my Denny T stage 2 fuel pin, not when I bought it anyway.
 
#29 ·
Because of your issue with elevation, I have seen where some have went tighter on the exhaust housing. for example you have the 30, take that wheel and housing off and put it on a 35. That exact set up may not be feasible I have seen where I may be able to run a 66/74/.90 here in Missouri but a friend in Idaho had to use a 66/71/.70 to get the same results
 
#31 · (Edited)


This will give everyone an idea of what elevation will do to a turbo compressor map. The temp is the same at 58.5* F (~15*C), the only difference is the absolute pressure going from standard (500ft) for the original map to my local absolute pressure which is 4500ft. You can learn how to do corrected flow from http://en.wikipedia.org/wiki/Corrected_flow . I just reverse engineered it so I could move the map to my elevation.

I wrote a program that will shift the maps according to abs. pressure and absolute temp, so that I can see how the high pressure turbo will react with certain boost and temp increases (adiabatic) from the LP turbo. I only have data for the hx30 and hx35, since those are the turbos I plan to use. I know this isn't the conventional way to do things with sizing but for me it seems easier to wrap my head around. Things may change in the future.

Next I will calculate the flow through the entire RPM range and plot it.
 

Attachments

#32 ·
View attachment 56586

This will give everyone an idea of what elevation will do to a turbo compressor map. The temp is the same at 58.5* F (~15*C), the only difference is the absolute pressure going from standard (500ft) for the original map to my local absolute pressure which is 4500ft. You can learn how to do corrected flow from http://en.wikipedia.org/wiki/Corrected_flow . I just reverse engineered it so I could move the map to my elevation.

I wrote a program that will shift the maps according to abs. pressure and absolute temp, so that I can see how the high pressure turbo will react with certain boost and temp increases (adiabatic) from the LP turbo. I only have data for the hx30 and hx35, since those are the turbos I plan to use. I know this isn't the conventional way to do things with sizing but for me it seems easier to wrap my head around. Things may change in the future.

Next I will calculate the flow through the entire RPM range and plot it.
You could also scale them in height.
 
#34 · (Edited)
Alright! Got some calculations done on what my compound set up will look like for now. I will need to do some tweaking later but one step at a time.

So I went ahead and did the calculations for a target torque curve similar to the stock 4BT that included the 600 ft lbs mark at around 2000 RPM as well as the 300 HP mark at 3200 RPM. Then I calculated the airflow that I would need to burn all the fuel for each of the points (A/F of 18) using the BSFC. Values of the BSFC above the 2700 mark, I estimated using the graph on the EPA tags. This torque curve starts at the same point at idle, then smoothly increases until 600 ft lbs then goes through the point that makes 300 HP at 3200RPM, I really only needed those two points since they were my only requirements, but I made the entire curve for approx target values through the entire range.

This graph is the airflow the engine needs inside the cylinder to produce that target torque curve and it also depicts the mass airflow (taking Volumetric Efficiency (VE) into account) that I need at the intake to get that much air inside the cylinders.

As you can see the mass airflow required at the intake gets significantly higher than the airflow required inside the cylinders, that is bc VE starts to drop dramatically with high RPMs. The difference between the two lines is what the VE will allow inside the cylinders. This is the mass airflow for basically everyone that wants to achieve this power level. For me it is going to take more boost to get there bc my absolute pressure is 12.5 psi at 4500ft.

Next I used the values that the HX30 and HX35 normally spool at and plotted them. I had to adjust the wastegating for each turbo until I could produce enough mass airflow to be at or above where I needed to be at the 2000 and 3200 RPM marks. I ended up wastegating the HX35 at 15 psi and HX30 at 25 psi. I put in about 3 psi loss in between each stage at rpm above 2000.
The HX30 normally spools to 25 psi by 2000rpm and the HX35 spools to 15 psi by 2500rpm (they probably spool faster but I am being conservative). I also have not included that the HP will also spool faster bc the LP is feeding it.

Then I calculate my total boost for my elevation and I am about 55 psi!!! (I used my PR ratios and my absolute pressure). Thats studs and o-ring territory. Hopefully, I will be running a smaller intercooler (if I can manage) between stages at 50% efficiency, then a large one after the HX30 (60%).

This is what my mass airflow will be with these boost levels compared to what my target line was:
My target mass airflow here is the same as the VE target airflow above, now I just compare it to what my turbos are expected to do.

So I will have sufficient air from peak torque to 3200 rpm. (I didn't factor in defueling). This mass airflow line will give me a different torque curve than the one I predicted.
Because my airflow will be less (theoritcally) at RPMs less than 2000, then I will have a little less power, but I will have more air than expected in the powerband 2000-3200rpm, so I can safely burn more fuel there if I wanted or I could leave my fuel and have the excess air cool my EGTs. This IMO is why Lonno's setup has such low EGTs, his twins are giving him a lot of air to cool the exhaust.

Next is to see if these values end up on my turbo maps. I know that at the higher RPMs the HX35 will definitely choke on the flow I need. And at that high of pressure and temp that the HX30 is seeing I may be getting close to surge. So I may have to 'up' the PR of the hx35 and lower the PR of the hx30 to stay on the maps. We will see...
 

Attachments

#35 ·
I don't yet understand how you calculated to those graphs (need to read and re-read), but they look good :D
 
#38 · (Edited)
Alright. Got some plots done on what the maps are looking like and we are gonna need to change things up a bit. With the settings described above, adiabatic compression, and dual stage intercooling (50% and 60%) this is what my temps are looking like:


The intercooler stage-1 outlet temp and the LP turbo boost (with piping psi loss) are what I used to move the hx30 map. Note: these graphs are for 4500 feet. I didn't want to plot graphs for each RPM/LP boost level and temp, so I combined the graphs in groups of 500 RPM segments. You can see how the maps move with each increase of 100 RPM.
I am just going to put the corrected flow in here instead of the groups of 500 segments. Easier to read and understand

At 3100 RPM we are hitting the top of the map of the HX30, anything above that we are off the map. This may work great when I factor in defueling though. Another thing that might help is if we let the LP turbo boost more and add more pressure to the HP turbo, this would shift the map more to the right and include more operating points.
Now let's look at the HX35 compressor map:

Looks like it will be pretty efficient from the get go but hits the choke line at about 2900 RPM. This is what I was afraid of. I don't know what size wheel goes to this compressor map. My guess would be that it is a 8 blade 54 or 56. So the 7 blades flow a little more, so i could go that route, or I could increase the PR of the hx35 and lower the PR of the 30 and see if I can get them both in more operating points. I probably will end up doing both since I am not worried about the hx35 hitting surge.
 
#40 · (Edited)
Yes. I make sure that when I do the corrected flow that the points lie proportionally to the original map as they do with the adjusted map and the high pressure/ high temps that I am working with. I just left them in the high pressure/high temp form so that we can see what is actually happening. For instance, I took the point on the HX30 flow at 3200. I need to flow 52.3 lb/min (the actual flow).

My inlet temp for the hx30 was 150.5 F = 610.5 degR, inlet abs pressure = 15psi boost + 12.5 (abs pressure) - 3 psi loss = 24.5 psia
Corrected to 58.5 F (518.5 degR) and 14.4 psia (29.4 in hg)

Corrected flow = (52.3 * (610.5/518.5)^0.5) / (24.5/14.4) = 33.3 lb/min

At a PR of 3, it is just barely outside the HX30 original map. Same as with my corrected map
 
#43 ·
Okay, I see, we're back to talking single turbos for the above graph. I thought you had switched to compounds.
In a 300hp compound set the corrected flow for a HX30 as the small turbo is about 20 lb/min.

Dougal, let me bounce this idea off you real fast. I have been thinking about defueling a lot because that will lower my PR and my flow at higher RPM, which is where things are getting tricky here. What if I bumped the max fuel up to 190 and started to defuel at around 2800rpm? From the boost numbers of the HX30 and HX35 (conservative side), it looks like I can get to 190cc/1000 at around 1800rpm (A/F = 18) and that is slightly over 600ft lbs. From there until defuel(2800) I can stay at 190cc/1000, thats when I can hit the 300hp mark (2800rpm), then defuel will happen and keep me inside the map for the hx35 for any rpm higher than that. My power band will be 1800-2800 and still reach the goals of this thread.

HX35 is the cheap way to go. Another way would be to get a S360, those I have seen for about $600 and I wouldn't have to worry about the choke like with these power goals.
I did a sea-level calc on the HX30-6 a while back, just looking at it again now.
I had capped boost at 32psi, power is about 190kw max and that happens around 2,500rpm and actually hangs pretty flat until around 3,200rpm. I was using 60% intercooling and 65% comp efficiency, you can better that with a good installation.

If you want to burn 190cc at 2,800rpm then you'll need about 37psi boost at 18:1 with 805 intercooling and 65% efficient compressor. Compressor flow around 36 lb/min (ask Alcaid what the 50mm HX30 can do) and power should be around 224kw (303hp).
 
#42 ·
Dougal, let me bounce this idea off you real fast. I have been thinking about defueling a lot because that will lower my PR and my flow at higher RPM, which is where things are getting tricky here. What if I bumped the max fuel up to 190 and started to defuel at around 2800rpm? From the boost numbers of the HX30 and HX35 (conservative side), it looks like I can get to 190cc/1000 at around 1800rpm (A/F = 18) and that is slightly over 600ft lbs. From there until defuel(2800) I can stay at 190cc/1000, thats when I can hit the 300hp mark (2800rpm), then defuel will happen and keep me inside the map for the hx35 for any rpm higher than that. My power band will be 1800-2800 and still reach the goals of this thread.

HX35 is the cheap way to go. Another way would be to get a S360, those I have seen for about $600 and I wouldn't have to worry about the choke like with these power goals.
 
#45 ·
So everything up until I hit the wastegate (when my PR's stop increasing) is still accurate, but when my flow goes way off to the right, is where things aren't gonna happen like that because the turbos aren't going to magically produce more mass airflow if the boost is not increasing.
In this quote, the last bit (emphasised) is not correct, because you are ignoring that the turbo speed will still increase as engine speed increases, up until the turbo chokes. You have the available flow indicated by the width of the map at any particular PR.
 
#46 ·
Right. I need to be more clear. I should have said "...magically produce A LOT more mass airflow...", meaning that it's not going match what is needed because I say so. It will produce according to RPM and VE. Sorry bush65, I'll fix that quote in red.
 
#50 ·
I think this works.

12.5psi atmospheric pressure.
HX35 at PR 1.8.
HX30 at PR 2.6.
Intercooler removing 60% of the heat.
BSFC of 242g/kwh but power reduced with extra pumping losses.
Mass flow 39.2 lb/min
Density increase 3.76:1
HX30 wastegate just cracked open. HX35 bypassing about 35%.
Drive pressure ~55psi.
Boost pressure~46psi.

Corrected flow for the HX30 is ~26lb/min.
Intake flow is a bit over 600CFM.
 
#51 ·
The mass flow of 39 lb/min at a PR of 1.8 for the HX35 is slightly outside the choke line at 4500 ft. That is also for 58.5 degrees F, so during the summer months, when its warmer, it will move it even more outside the map.
We need to raise the PR of the HX35 slightly to be included in more of the map. We could also lower the HX30 to match the total boost we need, but that would affect the bottom end response when the hx30 opens the wastegate. What is the downside to having more air than needed? Less heat energy for the turbines to extract? Having to stud and possible o ring? Anything else?
 
#52 ·
You'll be off the bottom end before the gate opens. You can tune both gates to dial in your desired PRs... You may want to limit how far the gate can open though to prevent unloading of the hx30 forcing the 35 to do all the work.

As far as too much boost, it can rob power as it takes power to make power... I would try to dial it all in while maintaining peak EGTs of around 1250F at full load which should net you close to a desired AFR
 
#54 · (Edited)
One thing I just thought of is instead of having a dual port actuator for the HX30, we could wastegate the hx30 to total boost at 44psi as Dougal has suggested elsewhere on this site, then the hx30 can hit a higher PR before it wastegates and we can move our powerband to lower rpm. Possibly 1700 for 600 ft'lbs with the cc/1000, I don't know, I'll have to ask Lonno how his spool.



If the 35 is gated at 14psi the hx30 couldn't unload, could it? It would have to make the desired total boost.

If your EGT's are really low bc of a high AFR, would that mean that you have too much boost and it is robbing power?
 
#53 · (Edited)
So I redid the calculations and I changed a few of the boost numbers up to make things work a bit easier. My boost and spool times are as follows:



I am wastegating the HX30 @ 20 psi and the HX35 at 14 psi. Boost of the HP and LP turbos are measured from directly outside the outlet. Total Boost is calculated using 3 psi boost loss after each stage. Using those numbers, using 50% and 65% intercooling on 1st and 2nd stages, and having the turbos operate at 65% eff, my maps show we will be well within the operating range on the HX35 and in the operating range of the HX30 but getting close to surge.




I converted my total boost flow into mass airflow to show how we are doing with the two points that we started out with: 600 ft lbs @ 2000 RPM and 300 hp @ 3200.



Looks like we got the air to make the power we want. That means I am getting pretty close to ordering parts that I don't have. But before I do that, I need to see what this setup will do going up into the mountains and taking a road trip to the beach. Also I will fluctuate the temps at each of those locations to make sure the turbos will work properly under each circumstance.

Keep bringing up problems that I might run into, so that I can address them with my build
 
#56 · (Edited)
Huh. I calculate ~45.1 lb/min for the corrected flow and that is outside the map that I am using (Midrange Outline Compressor Maps) for Holsets. We could be using different maps for the HX35 though.

Cor_flow = Actual * sqrt(T_in/T_map) / (P_in/P_map)

Ref temps and pressure
29.4 in Hg = 14.4 psia
58.5 F = 518.5 degR

Cor_flow = 39.2 * sqrt(518.5/518.5) / (12.5/14.4) = 45.15 lb/min

That is right off the edge of choke. Anyway, I would like to stay away from that edge if I can, if I go any higher in altitude, I would need to be sure my foot is off the pedal. If I have a small intercooler between stages that would keep the temps down, but at the same time shifts my corrected flow to the left toward choke. Cheap compounds is walking a fine line. haha.

Anyone have some experience on how fast these turbos will spool in compounds?
 
#57 ·
I've got two minor differences in my calcs.
I'm using 86.1kPa for actual pressure (12.5psi) and 96.3kPa (13.9Psi) for reference pressure. I can't remember right now where the 96.3kPa came from. But I will have it referenced somewhere. Does it match the holset maps correction factor? I haven't had time to check.
I also have a small temperature difference which is coming into play. 303K for reference temp, 293K for actual temp.

So that's why we're getting slightly different corrected flow numbers.
As for flow, we're using the same charts, the yellow outline runs to about 44 lb/min at PR 1.8. There is also a blue outline for which the label includes HX35 which goes a lot further.

Regarding spool. I think you won't hit target boost until ~2,500rpm. Power is looking relatively flat from 2500-3200rpm.
You'll have ~600Nm from ~1500rpm on up. Peak torque around 2,500rpm.

But these are unpolished results. My compressor efficiencies aren't right and turbine efficiency is a stab in the dark too.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top