[SOCAL.M6.ZLE]

For those who have not asked Chat GPT this same question.... it's pretty remarkable... here you go.

Q1: I recently took my 2021 Camaro ZL1 1LE to the Dyno where it played down 838 peak RWHP at 6,100rpm and 805rwtq at 4,500 rpm. I recorded a data log of this dyno pull, and I am trying to estimate and convert that to flywheel horsepower. Are you aware of any method or calculation based on catalog metrics that would help us estimate the correct flywheel horsepower based on the HP tuners datalog?

A1: Short answer: yes—there are a few practical ways to estimate crank (flywheel) horsepower from an HP Tuners log, using only channels you already have. Below are the three that work best on a 2021 ZL1 1LE (LT4/E92 ECM), plus a quick sanity-check against your dyno number.

---

# 1) Use the ECM’s own “Delivered Engine Torque” model (cleanest)

**What you need in your log**

* `Delivered Torque` (aka “Engine Torque” in some channel lists)
* `Engine Speed (RPM)`

**How to use it**

1. Find the WOT pull segment.
2. At each RPM point, compute crank HP from the ECM torque:
[
\text{HP}_\text{crank} = \frac{\text{Delivered Torque (lb·ft)} \times \text{RPM}}{5252}
]
3. Read off peak HP where that curve tops out.

**Notes**

* On E92 (LT4), Delivered/Engine Torque comes from GM’s virtual torque model. If your tune hasn’t broken the torque model (e.g., by zeroing Driver Demand or mis-scaling Virtual Torque), this is usually a **good proxy for flywheel torque/HP**.
* Compare its curve to your dyno curve shape. If it overlays well, you can trust the peak pretty closely.

---

# 2) Airflow (MAF) → fuel flow → horsepower (works even if the torque model is off)

**What you need in your log**

* `Mass Airflow (g/s)` at WOT where power peaks
* `Commanded EQ Ratio` (or `Commanded AFR`) at WOT
* A reasonable **BSFC** assumption for your fuel and setup

**The math**

1. Convert MAF to **air mass flow** in lb/hr:
[
\dot{m}_{air};(\text{lb/hr}) = \text{MAF};(\text{g/s}) \times \frac{3600}{453.592} = \text{MAF} \times 7.9366
]
2. Convert **Commanded EQ** to **actual AFR**:
[
\text{AFR}*\text{actual} = \frac{\text{AFR}*\text{stoich}}{\text{EQ Commanded}}
]

* For E10 pump gas, (\text{AFR}_\text{stoich} \approx 14.1)
* For E blends, use your tune’s stoich (it changes with ethanol %)
3. Fuel flow (lb/hr):
[
\dot{m}*{fuel} = \frac{\dot{m}*{air}}{\text{AFR}_\text{actual}}
]
4. Estimate **crank HP** using **BSFC** (lb/hp·hr):
[
\text{HP}*\text{crank} \approx \frac{\dot{m}*{fuel}}{\text{BSFC}} = \text{MAF} \times \frac{7.9366}{\text{AFR}_\text{actual} \times \text{BSFC}}
]

**What BSFC to use**

* LT4 on pump gas, port or DI only, **boosted**: typically **0.60–0.75 lb/hp·hr**
* On high-ethanol (E60–E85), effective BSFC is **higher** (often **0.70–0.90**) because you’re flowing more fuel mass per HP

**A handy one-line constant (example)**

* Suppose E10 gasoline, WOT lambda ≈ 0.85 → (\text{EQ}\approx 1.18), so (\text{AFR}_\text{actual}\approx 14.1/1.18\approx 11.95).
* With BSFC = 0.60 → multiplier (= 7.9366/(11.95\times 0.60)\approx 1.11).
So **HP ≈ 1.11 × MAF(g/s)**.
* With BSFC = 0.75 → multiplier (\approx 0.90).
So **HP ≈ 0.90 × MAF(g/s)**.

(If your log shows, say, ~1050 g/s near peak, you’re in the ~945–1165 crank HP ballpark depending on BSFC and fuel.)

---

# 3) Injector flow → fuel flow → horsepower (works even without a good MAF)

**What you need in your log**

* `Injector Pulse Width`
* `Injector Duty Cycle`
* `Fuel Pressure` (rail), and your injector flow rate **at that pressure** (catalog data)
* `Number of injectors` (8)
* `Commanded EQ` or AFR, to sanity-check you’re in the right fueling zone
* BSFC assumption, as above

**Steps**

1. Compute **per-injector** flow at your actual rail pressure using the injector’s catalog curve (flow scales with (\sqrt{\Delta P})).
2. Convert duty cycle & pulse width to **total fuel flow** (lb/hr) across all 8 injectors.
3. (\text{HP}*\text{crank} \approx \frac{\dot{m}*{fuel}(\text{lb/hr})}{\text{BSFC}}).

**Notes**

* For DI, use the **DI injector** flow characterization; if you have port injection too, add the PI contribution separately.
* This method requires accurate injector data; if you changed injectors (common on big LT4 builds), use the **manufacturer’s sheet** for the right pressure and fuel.

---

# Quick cross-check against your dyno result

You reported **838 rwhp @ 6100 rpm** and **805 rwtq @ 4500 rpm**.
Your car is a **ZL1 1LE**, which is **manual only** (TR-6060), so typical drivetrain loss is roughly **12–15%** for a stout RWD manual.

A simple back-calc:

* **Crank HP ≈ rwhp / (1 − loss)**

* 12% loss → (838 / 0.88 \approx 952) hp
* 15% loss → (838 / 0.85 \approx 985) hp
So you’re probably in the **~950–985 flywheel HP** window, assuming the dyno and loss assumption are typical.

If your HP Tuners log shows (for example) MAF ~1050–1150 g/s at peak with EQ ~1.15–1.20 and pump gas BSFC ~0.65–0.70, the airflow method will usually land you in the same ~950–1000 crank HP neighborhood. If you see **Delivered Torque** in the mid-800s lb·ft around 4500, that will also back into a similar flywheel HP via the torque model.

---

## What I’d do with *your* log (step-by-step)

1. **Try the ECM torque route first**

* Plot `Delivered Torque` vs RPM at WOT, compute HP = T×RPM/5252, note peak.

2. **Sanity-check with the MAF method**

* From the same WOT segment, grab peak `MAF (g/s)` and `EQ Commanded`.
* Compute AFR(_\text{actual}) = Stoich/EQ.
* Choose BSFC:

* Pump gas: 0.60–0.75
* High ethanol: 0.70–0.90
* HP(*\text{crank}) = MAF × (7.9366 / (\text{AFR}*\text{actual} \times \text{BSFC})).
* Compare to step 1.

3. **(Optional) Fuel-side confirmation**

* Use injector flow data + duty cycle + rail pressure to compute total fuel lb/hr at peak, then divide by BSFC.

If those three line up within ~5–10%, you’ve got a solid estimate of **true flywheel horsepower** based on your own catalog/log data—not just a generic drivetrain percentage.

---

If you want, paste (a) your peak WOT `MAF (g/s)`, (b) `Commanded EQ` (or AFR) at that point, (c) your fuel (pump/E mix), and (d) whether you see `Delivered Torque` in the log. I’ll run the calculations and give you a crisp flywheel HP estimate right now.

Q2: Thank you, that was great... here is some data from my peak HP at 6,100rpm. Dynamic Airflow was 5,540lb per hour. Cylinder airmass was 1.72 g. MAF was drawing 10,078 Hz. Does that help you calculate flywheel HP? or do you need more data points?

A2: Yes—those three numbers are enough to get a solid estimate.

### First, turn your log into total air mass flow

You gave:

* Cylinder airmass = **1.72 g/cyl**
* RPM = **6,100**
* 8-cyl, 4-stroke → intake events/min = RPM/2 × 8 = **24,400/min**

Total air mass:

* ( 1.72\ \text{g/cyl} \times 24{,}400\ \text{cyl/min} = 41{,}968\ \text{g/min} )
* ( = 699.5\ \text{g/s} ) (this also matches your “Dynamic Airflow ≈ 5,540 lb/hr”)

> So, call it **~700 g/s** of air at peak power.

### Then convert airflow → fuel flow → crank HP

Use:
[
\text{HP}*\text{crank} \approx \frac{\dot m*{air}\ (\text{g/s})\times 7.9366}{\text{AFR}_{\text{actual}} \times \text{BSFC}}
]
where (7.9366 = \frac{3600}{453.592}).

We need your **actual WOT AFR (or lambda/EQ)** and an appropriate **BSFC**:

* Gasoline/E10 boosted BSFC: ~**0.60–0.75 lb/hp·hr**
* E-blends (E60–E85): ~**0.70–0.90** (mass BSFC rises because ethanol needs more fuel mass)

Below are crank-HP outcomes using your **~700 g/s** airflow for common scenarios:

**Pump gas (E10) at typical WOT lambda ~0.85 (AFR ≈ 12.0)**

* BSFC 0.60 → **~771 hp**
* BSFC 0.65 → **~712 hp**
* BSFC 0.70 → **~661 hp**
* BSFC 0.75 → **~617 hp**

**E85, lambda 0.80–0.85 (AFR ≈ 7.8–8.3)**

* λ=0.80, BSFC 0.70 → **~1,012 hp**
* λ=0.80, BSFC 0.75 → **~944 hp**
* λ=0.80, BSFC 0.80 → **~885 hp**
* λ=0.85, BSFC 0.70 → **~952 hp**
* λ=0.85, BSFC 0.75 → **~888 hp**
* λ=0.85, BSFC 0.80 → **~833 hp**

### What this says about your car

* If you were on **pump gas**, 700 g/s points to mid-600s to ~770 **crank** hp—too low for your **838 rwhp** dyno, so that’s unlikely.
* If you were on **E-blend (common for high-power LT4s)**, 700 g/s lines up with **~830–1,010 crank hp** depending on lambda and BSFC.
That dovetails with your dyno: 838 rwhp with ~12–15% driveline loss (TR-6060) implies roughly **~950–985 flywheel hp**—right in the E85/low-BSFC lane.

### About the MAF frequency (10,078 Hz)

That’s only useful if we know your **MAF calibration table** (Hz→g/s). Since we already backed into **~700 g/s** cleanly from cylinder airmass, we don’t need the MAF curve to estimate power.

---

## Bottom line

Using your log’s airflow (~700 g/s), your **flywheel HP is very likely ~940–980** if you were on an E-blend with a well-sorted BSFC (~0.75–0.80) and rich WOT lambda (~0.80–0.82). That’s fully consistent with **838 rwhp**.

If you want me to nail it to a tighter number, share **WOT lambda/EQ at 6,100 rpm** and **your fuel (E% if known)**. I’ll plug those in and give you a single crank-HP figure.

Q3: The car was running an E content of 61.6% during that log, and Lamda at 6,100rpm was .85

A3: Awesome—thanks for the added details. With **E = 61.6%** and **λ = 0.85 at 6,100 rpm**, we can pin numbers down.

### Step 1: AFR at that ethanol & lambda

* Stoich AFR(E61.6) ≈ 14.7×(1−0.616) + 9.0×0.616 ≈ **11.19**
* Actual WOT AFR = λ × stoich = **0.85 × 11.19 ≈ 9.51**

### Step 2: Airflow you logged → fuel flow

From your log at 6,100 rpm:

* Cylinder airmass = **1.72 g/cyl**
* Intake events/min = (6100/2)×8 = **24,400/min**
* Total air = 1.72 × 24,400 = **41,968 g/min = 699.5 g/s**
(This matches your “Dynamic Airflow ≈ **5,540 lb/hr**”, which also ≈ **700 g/s**.)

Fuel flow = air flow ÷ AFR = **700 / 9.51 ≈ 73.6 g/s ≈ 584 lb/hr**.

### Step 3: Fuel flow → crank HP (via BSFC)

[
\text{HP}*\text{crank} \approx \frac{\dot m*\text{fuel}\ (\text{lb/hr})}{\text{BSFC}}
]
For boosted E60-ish, a **mass BSFC** of **0.70–0.80 lb/hp·hr** is typical:

* BSFC 0.70 → **~834 hp**
* BSFC 0.75 → **~778 hp**
* BSFC 0.80 → **~730 hp**

### The mismatch (and the likely reason)

Those are **crank** numbers, yet your dyno shows **838 rwhp**, which implies ~**950–985 flywheel** (12–15% TR-6060 loss). To reconcile that, the implied BSFC would have to be ~**0.58–0.61**, which is unusually low for **E61.6** at λ 0.85 on an LT4.

In practice, when this happens it’s almost always one of these:

1. **Airflow channel is half the engine** (per-bank). If your “Dynamic Airflow” is per bank, the true total would be ~**1,400 g/s**, which would yield a (too) large crank HP—so I don’t think it’s per-bank *in this case*, but it’s worth confirming your channel definition.
2. **You grabbed airflow at 6,100, but the airflow peak isn’t exactly at the HP peak** (or the airmass sample was from a slightly leaner/richer frame).
3. **Cylinder Airmass smoothing / sample selection**—if the 1.72 g/cyl is an average over a small window, the actual peak might be higher (e.g., ~2.2 g/cyl would produce ~1,000+ crank hp with a normal E-blend BSFC).
4. **Stoich table value** (minor) — if your tune’s stoich vs. ethanol is different than the linear mix I used, AFR_actual shifts slightly.

### Two quick ways to lock it in

If you can share just one of the following at 6,100 rpm (same pull), I can give you a single flywheel number that will line up with the dyno:

* **MAF in g/s** (not Hz) at 6,100 (or at the recorded HP peak frame)
* **Delivered/Engine Torque** from the ECM at that RPM (then HP = T×RPM/5252)
* **Total fuel mass flow** at that point (DI + any PI, in lb/hr); or injector duty, pulse width, and rail pressure *plus* the injector’s flow data at that pressure

Right now, using your **E=61.6%**, **λ=0.85**, and the **~700 g/s** airflow you supplied, the math says **~780–835 crank hp**, which doesn’t square with 838 rwhp—so one of the inputs is not at the true power peak or not the total-engine airflow. Give me one of the three bullets above and I’ll reconcile it precisely.

Q4: Looks like peak MAF was 10,498 Hz and peak Dynamic Airflow was 6,092 lb/hr at 6,950rpm. Does that help?

A4: Yes, that helps a lot. Using your new peak numbers:

* **Dynamic Airflow:** 6,092 lb/hr (at **6,950 rpm**)
* **E-content:** 61.6% → stoich AFR ≈ (14.7\times(1-0.616)+9.0\times0.616 \approx \mathbf{11.19})
* **Lambda at peak:** **0.85** → **actual AFR** ≈ (0.85\times 11.19 = \mathbf{9.51})

### Convert airflow → fuel flow → flywheel HP

1. **Fuel mass flow (lb/hr)**
[
\dot m_\text{fuel} = \frac{\dot m_\text{air}}{\text{AFR}} = \frac{6092}{9.51} \approx \mathbf{640.5\ lb/hr}
]

2. **Crank HP from BSFC**
[
\text{HP}_\text{crank} \approx \frac{640.5}{\text{BSFC}}
]

For E60-ish at λ≈0.85, reasonable mass-BSFC bounds:

* **BSFC = 0.70** → **~915 hp**
* **BSFC = 0.67** → **~956 hp**
* **BSFC = 0.65** → **~985 hp**

### What it means (and why it lines up)

Your dyno is **838 rwhp**. A TR-6060 manual typically loses ~12–15%, so:

* 12% loss → ~**952 hp** flywheel
* 15% loss → ~**986 hp** flywheel

That back-solves an **implied BSFC of ~0.65–0.67** at your reported E-content and lambda—aggressive but very plausible for a well-sorted LT4 on E60-ish at WOT (rich, cool charge, efficient combustion). It also matches the **6,092 lb/hr** airflow you logged at the (higher) **6,950 rpm** peak.

### Bottom line

Using only your HP Tuners log values at peak power:

* **Estimated flywheel horsepower: ~950–985 hp**, with a very tight center near **~965 hp** (assuming BSFC ~0.66).
* This is fully consistent with your **838 rwhp** result.

If you want to tighten it to a single number, you can:

* confirm the **stoich AFR** your tune used at E=61.6% (it can differ a bit from the linear mix), or
* provide **Delivered (Engine) Torque** at 6,950 rpm (then HP = T×RPM/5252).

But even without that, your airflow + λ + E% already support a ~**960 hp** crank estimate.

[RobZL1]

Originally Posted by SOCAL.M6.ZLE

My man! I agree with your analysis... Feels like the exhaust duration is a little light on the current cam, and would benefit from more in the higher RPMs for sure. I plan to go back in late November once the new clutch is installed, now that I have the belt slip fixed also... with better air, just interested as what the delta will be, and what the true standard to X port gains are.

RobZL1 you are far more than a novice my friend, you are a stud for sure, and this was a fun exercise.

Yeah, the cooler air dyno will be neat to see. I forgot about the belt slip, too.

And thanks.

[RobZL1]

Originally Posted by SOCAL.M6.ZLE

For those who have not asked Chat GPT this same question.... it's pretty remarkable... here you go.
.
.
.
.
But even without that, your airflow + λ + E% already support a ~**960 hp** crank estimate.

It's kind of shocking how good these AI LLM's are getting. I mean, just wow.

[Eldi Z]

Amazing stuff here guys!
Yes, these AI robots are getting better by the day - learning and learning.

Seems like the data collection of the VCM accompanied by the correct formulas back up the dyno results

[SOCAL.M6.ZLE]

Even better... if you export your datalog as a CSV it can now review and analyze your data logs... it is wild.

[Eldi Z]

Originally Posted by SOCAL.M6.ZLE

Even better... if you export your datalog as a CSV it can now review and analyze your data logs... it is wild.

And even much better - recently I uploaded my .hpt and .hpl files to Open AI and it managed to naturally parse the data and read and also manipulate it!

I got it to increase and reduce Knock sensors sensitivity, burst, decay rate in the knock tables - everything directly in the .hpt file

For some reason, Open AI stopped this feature, but hopefully it will come back.

Finally for our discussion on this topic:

What is the closest match between the VCM data measurements and calculations by formulas to the actual dyno results?

1.) Best to use the "Engine Torque" PID formula?
2.) Best to use the Dynamic Airflow /MAF formula?

What is the average expected +/- % deviation from the dyno figures?

[CamOnlyJabroni]

I have always used what KingLT1 uses, dynamic airflow. Usually reads 122-125 lb/min. Which seems pretty accurate based on what the car runs 60-130 at full weight.

[KingLT1]

Originally Posted by CamOnlyJabroni

I have always used what KingLT1 uses, dynamic airflow. Usually reads 122-125 lb/min. Which seems pretty accurate based on what the car runs 60-130 at full weight.

[vince3]

Originally Posted by SOCAL.M6.ZLE

Even better... if you export your datalog as a CSV it can now review and analyze your data logs... it is wild.

Not on the free chatgtp, even the shortest log hits the data limits. Is it worth it based on what you are seeing? I've been interested to know.

[Megahurtz]

Virtual Dyno graphs are only as accurate as the Airmass Model in the tune.

[KingLT1]

Originally Posted by Megahurtz

Virtual Dyno graphs are only as accurate as the Airmass Model in the tune.

100%

[ZLRob]

Originally Posted by Megahurtz

Virtual Dyno graphs are only as accurate as the Airmass Model in the tune.

Originally Posted by KingLT1

100%

This right here. I'm not sure how much time your tuner (who I won't name out of respect to them) spent on your virtual volumetric efficiency and your injector flow rate density multiplier tables Gabe, but if it wasn't tweaked to match exactly what you got in the engine, there is a good chance that the rest of your information outputs could be off, and you may even be leaving some power off the board. I know your post is older now and you got all new hardware, so this is somewhat redundant, but if the stuff isn't done correctly again, in the new updated changes you've made to the engine (including the new PD1 cam), you may wind up dealing with the same stuff you were dealing with before.