Vaporworx / Rick's Tank Problem - dropping fuel pressure at high rpm?
 

Tonight while running through some datalogs from my tuning runs this afternoon I noticed something strange. My fuel pressure at the fuel rails is dropping at mid-to-high RPM. I did not expect to see that, so I'm wondering if this would be considered 'normal' or not.

I have a Rick's Tank with the Vaporworx setup, with the (I think) 5th Gen fuel pump / regulator as purchased from Rick's. I'm running a combination of -6AN and 3/8" hardline from the tank to my fuel rails, and my pressure sensor is at the dead-head end of the rail. I'm running a set of 39 lb/hr injectors on the engine.

You can see in the plot below that the AFR as measured is pretty stable, so perhaps my ECU has already 'tuned in' the drop in pressure. But I was surprised to see a 6# drop.

Has anyone else noticed this with the same pump / tank setup? Anyone else surprised by this?

BTW, I only noticed this because, of all things, my ignition timing dropped way down at this time, and I couldn't figure out why. I then realized that the Holley HP EFI setup I have was pulling timing (as I have it set up to do) when the fuel pressure dropped below 53psi. So at least I 'tested' out that functionality... ;)

http://i100.photobucket.com/albums/m...ps8b78fc9c.gif

I am running the same Vaporworx system on my car, other than I have the CTS-V fuel pump rated at 380lph. I also have a 6AN line running from the fuel tank to the fuel rails. I am running my system pressure at 58psi and it never varies more than a couple of psi from idle up to peak rpm. My Vaporworx pressure sensor is mounted in-line very close to the fuel tank. I measured the system pressure at the fuel rail with a mechanical gauge.

I know the 5th generation Camaro LS3 fuel module is rated at 255lph, so depending on your HP level maybe you need more capacity on the pump? I don't think it's normal to drop 6psi, mine stays very consistant at 58psi.

Thanx for the feedback. This engine was dynoed at 615HP when built, but since I'm still tuning, I don't think I'm up in that range at this point.

And the strange thing to me is that it starts losing pressure at lower RPM. I'm down 2PSI at 3500 RPM when I did the 1-2 shift, and 3PSI at 4000.

I'm going to need someone with more engineering knowledge about the flow requirements than I have off-hand to help determine if I need a bigger pump. ;) I'll try to Google it later, but I would be surprised if I'm losing pressure due to lack of flow that early in the engine speed scale.

I'll see if I can add the injector flow rate to the chart. Maybe that will signal something to me as well...

Just for reference... I tested the Vaporworx digital controller and fuel pump when I ran my engine on the engine dyno. My engine put out 672HP @ 6,900 rpm... I measured the fuel pressure and voltage output of the controller. At 1,200 rpm the fuel pump digital controller measured about 7 volts to the fuel pump, and at 7,200 rpm the controller measured between 11 to 12 volts to the fuel pump. I had plenty of capacity to support my engine with the 380lph pump... but I don't think the the 255lph Camaro pump would have been enough.

My fuel pressure does drop slightly for an instant when you first hit it but no more than 1 or 2 psi at any time.

So I tested a bit more this afternoon, and I saw a full throttle run where I dropped to 45 psi! That's no good...

I discovered that the Holley HP EFI will actually give me a Fuel Flow calculation as one of the pieces of data in the Datalog. If the number is to be believed, then it showed a max value of 299.2 lb/hr. If your 5th Gen pump number of 255 lb/hr is right, and my Holley number is right, then I'm out of fuel pump. :(

That would seemingly make sense, but it's also a bit surprising to me. I need to go back and reread the Vaporworx threads, but I thought the 5th Gen pump was the recommended setup. Maybe it's only good up to about 500HP or so?

Bryan - like Dale I've got the CTSV pump. I believe the 5th gen is maxed out at 600hp, and in the real world maybe a little less. IIRC, my injectors were spec'd a little larger than yours as well.

Get in touch with Carl and he'll get you straight.

Bryan, If you are still testing today try this... measure the voltage output of the digital controller (12V positive output to the fuel pump) and if it measures 12 volts before you get to peak rpm then you know the fuel pump is not large enough. From my understanding... the digital controller increases the voltage to the fuel pump as the demand for fuel goes up- to maintain the desired fuel pressure. If you measure 12V output at the controller before peak rpm then your out of pump capacity.

Carl helped me to size my fuel pump and I believe that he said when you get around 600HP then you need the CTS-V pump like Ron said. Carl is a great resource for help. Also.. the capacity on the 5th gen pump was 255 liter per hour rather than lb/hr.

For some reason I keep thinking that the fifth gen pump should be able to handle your amount of power. When I spoke with Carl I remember debating which pump to run and choosing the CTSV because because the fifth gen pump wouldn't be able to support the engine if I got above 650-680ish. I think that is where it maxes out on n/a engines. Gregg B is running a fifth gen pump to its limit and has around 560rwhp. He hasn't mentioned a drop in pressure at high rpm and he is pretty careful about getting everything right.

I had a similar conversation with Carl about the max horsepower limits the 5th gen module could handle and he said he had a customer using one on a motor with 630hp. I hope Carl chimes in soon as I am contemplating some mods and would like to know the limits of the 5th gen module.

I'm actually not using the Digital Controller, I'm just using the Gen 5 pump assembly and its controlled by the HP EFI. So it's either on or off, I'm not using the PWM controller.

I've been re-reading the Vaporworx threads, and I think I'm going to order one of the controllers. But I don't know that it will solve my problem, since I'm running it full blast already.

I captured some more data last night that I need to process, I'll try to post up another screenshot later.

Hmm, I need to do a conversion then to see how the liter/hour converts to the lb/hr value. It sounds like I'm on the edge of what the 5th Gen pump can handle though, although I would think that given that I'm not really fully tuned yet, I wouldn't think I'm running full fuel. Perhaps I have it set a touch too rich, and maybe leaning it out will help, but if I'm that close, then it's a problem it sounds like...

I feel like it should handle it also. It's strange to me that I start seeing the pressure drop at lower RPM, where I'm certainly not making that kind of power. I checked the battery level (via the Datalog) during the run, and there wasn't a drop in voltage, so that wasn't the problem either...

How close to the battery did you mount the controller? I'm wondering if there might be something not working properly between the battery/controller/pump.

Matt,

The ECU is mounted on the cowl in the cab, if that's what you were asking. I don't have the Vaporworx PWM controller (yet).

If the conversion factors that I'm finding for lb/hr to l/m are correct, then at 300 lb/hr (which my ECU reports), that should be 3.15 lpm. Multiplied by 60 m/h, that should only be 189 lph. So I should be within the limits of the 5th gen pump. Unless my math is wrong, and obviously I'm just relying on the ECU data.

Sorry guys, I'm just finding this one.

The PWM controller won't increase output. It slows down the pump to decrease the amount of heat generated and to reduce wear. The only thing that is going to increase pump performance is to increase the input voltage.

The Gen5 LS3 pump was used on James Shipka's OLC with the Katech LS7 was a nice match for that setup. It made about 630hp. The pump had a small safety margin in that car but it supplies 14 volts at the controller input. That helps pump performance. Engine efficiencies and input voltages can make a big difference here as well since it's getting close to the maximum output of the pump.

It is normal that pressure will drop in a non-manifold referenced system during part throttle. The C5 Corvette fuel filter/regulator does the same thing and has very similar output curves. As more fuel flows to the engine, the pressure drop will increase to the point where no more fuel is bypassed by the regulator. There is another pressure regulator available that does not have as much pressure reduction, Delphi FP10075. It does have a slightly higher it-idle pressure of 61-62psi but will only lose about 3psi at full bypass.

Three things come to mind here:

One, the fuel pressure regulator may be partially dislodged and causing an internal leak. To check for this fuel module removal is required.

Two, there may be a restriction in the fuel line. To check for this a fuel pressure gauge would need to be placed near the fuel module outlet and compared to the pressure at the injector inlet. A small difference will be normal as there are losses due to friction and fitting restrictions.

Three, the pump may be too small. At 50psi the regulator should be sending all of the fuel to the engine. At 50psi/13.5v the pump can supply 326 #/hr, enough to feed a naturally aspirated engine to 652 horsepower at a conservative 0.5bsfc with an appropriate injector. However, there's not much safety margin.

Some numbers:

At 13.5v:

60psi = 300#/hr or 189 liters/hr
55psi = 313 197
50psi = 326 206
45psi = 339 214
40psi = 353 223

For naturally aspirated engines the brake specific fuel consumption (bsfc) is considered to be 1/2# per hour of fuel for each horsepower. So, at 50psi the pump can provide 326#/hr. Divide this by 0.5 (1/2). That results in 652hp. However, there's no safety margin.

You may also want to check here for some injector calculations http://www.rceng.com/technical.aspx

For 650hp, 80% injector duty cycle, and 58psi fuel pressure a 44 lb/hr injector would be a good choice. It sounds like you may already be at the limits of your current injector as well, but the fuel pressure side of things should be fixed first.

Injectors are typically rated for 43psi (3bar), not 58psi (4bar). I won't go into the gory details of all this unless you want me to, but it's a good practice for us. So, at 42psi you're looking at a 52 lb/hr injector (assuming eight injectors.)

You may also want to check that the wire gauge to the relay(s) is 14ga minimum and that there is not a large voltage drop to the relay output. The Gen5 LS3 pump needs 12 amps.

If it is determined that the fuel pump needs to go bigger the easiest thing to do would be to put in a ZL1 fuel module. You can test this by putting in the fuel module using the 58psi regulator that is in your current fuel module. Though not recommended for extended use, it will work to get the system figured out. The ZL1 is capable of:

60psi = 391#/hr = 247 l/hr
55psi = 415 262
50psi = 436 275

It also needs 17 amps, so a PWM system is recommended for this setup for long-term use.

Carl,

Thank you for responding with so much information! Of all of the suggestions that you have provided, I think the following may be the one I need to focus on:

I'm using the 'relay' from the Holley HP EFI kit, which supposedly will source 10A. I do have 12 or 14ga wire, I don't recall which at the moment. I unfortunately do not have a good way to measure the actual voltage at the pump itself, but if it requires 12A, then it's very possible that I have a limitation on the available current. I did not run an extra relay for the pump itself, since I was under the impression that the HP EFI could source it. That may be my mistake.

If I'm going to dig back into this wiring, then it may be a good time for me to go ahead and install your PWM controller instead of just a relay. I don't think it was available when I made this original purchase, which is why I don't have one now...

I will contact you via PM about acquiring one...

Thanx again!

I'm glad to help Bryan.

If you want to to a check of the wiring remove the circuit that connects the relay to the fuel pump. Take the output from the relay and run it to BAT + (assuming it's +). Do the same for the negative, or confirm that the chassis ground is secure and has direct contact with the chassis material, i.e. no paint, corrosion, etc. It won't hurt the pump to connect it like this with the engine not running. Connect everything with the engine off, then try running the car again. Use an in-line fuse to protect the circuit. The circuit should be capable of 20A in either case.

In the end you may well find out that the LS3 pump is a bit small for the build. The ZL1 has almost identical form, fit, and function to the LS3, the only difference being that the ZL1 won't compress quite as much as the LS3 due to the taller pump in the module reservoir. This is only a factor if you have retro-fitted an older Rick's, Rock Valley, or stock tank since these are only 7" tall overall. All of the tanks fabricated by Rick's or RV have sufficient mouting height to accomodate either fuel module.

The basics of the PWM kits will interchange among the LS3, ZL1, an CTS-V single fuel modules. The LS3 and ZL1 kits are identical. The CTS-V has a different short harness since the wiring for the plug on the fuel module hat is different. In other words, if you have a PWM kit for an LS3 and want to use a ZL1, just put the new pump in and go. When running twin fuel modules, like that in Mark Stielow's latest builds, it requires a different controller tune to get the system happy when running both pumps at the sam time, plus there is a lot more wiring to tie in the second module.

Hey carbuff, I was directed to this thread by a community member, I work for a fuel pump and injector manufacturer and was actually in the process of flow testing the stock LS3 pump from the 5th Gen Camaro.

From 30psi up to 100psi, it was using 10.6-15.2 amps @ 13.5 volts. If I'm reading your first sentence right, you are over extending the Holley EFI relay.

From the looks of it, the pumps is actually quite good. My numbers are showing for 675hp at the crank you'll need around 195LPH of fuel from the pump. At 60psi, the LS3 pump was putting out 220LPH by our testing equipment.

As Carl pointed out, you'll need around at least 45# injectors to keep the IDC's in the 80's with 675hp...and also a 44# injector at 58 psi will yield ~52#.

If your current injectors are essentially going static, I could see that being a source of your fuel pressure drop, as well.

If air/fuel ratio is correct that will not have an effect.

Low,

Thank you for chiming in! Always good to have someone 'in-the-know' to offer help and information on these forums!

That seems to be the consensus. It is interesting to me that it will hold the 58PSI consistently though until I actually open the throttle and the RPM start to rise though. You can see that on the left and right of the picture below (the pink line).

That's obviously more PSI than I was seeing, but my measurements were after the injectors, so I don't know if we are actually comparing apples to apples...

This touches on a subject that I need some education on, so anyone who can help, I would appreciate it!

My car currently has 39# injectors. As best I can tell, that rating is at 3 BAR, or 43.5PSI. I'm running my system at 58PSI (at least when it keeps up), so using the formula which I've found online, that converts to about 45# equivalent.

So your statement above, that I need 45# injectors, does that mean that I have them? Or are you saying that I need 45# injectors, rated at 3 BAR?

I attempted some math to determine if I am going static or not. What I came up with is this. At 6500 RPM, each revolution takes about 9.2 mS, and 2 revolutions takes 18.4 mS. I'll paste another screenshot from the Holley Datalogger below. It's telling me that the Injector Pulse width is running consistently in the range of 16.0 - 16.3mS. If that, and the above 18.4mS at 6500RPM is correct, that would be about 87% duty cycle at 6500rpm.

Also note in the new picture below that my actual AFR is hanging at almost exactly what my Target AFR is (after recovering from Acceleration Enrichment, caused by the rate-of-change in my TPS and MAP), which was set to 12.5:1 for this run. This is also the run which shows the rail fuel pressure dropped to 44PSI.

I find several things about this interesting. First, the Injector Pulse width stays almost constant throughout the RPM range. Maybe that makes sense if I'm requesting the same AFR at all times (which I believe I am, and the Target AFR line appears to show).

So... Do I need to increase my injector size? Based on AFR, they seem to be keeping up. And do I need to keep worrying about this, if my AFR is keeping on target?

I have decided to install one of Carl's Vaporworx PWM controllers, if for no other reason than to get a clean voltage to the pump that isn't limited by my ECU. I'll be curious to see what differences that makes on my readings!

Thanx!

http://i100.photobucket.com/albums/m...ps60d065c5.gif

That statement was you need 45# worth of fuel coming out of the injectors to be in the 80% duty cycle range. I know the whole nominal flow vs. actual flow thing can be a little confusing.

I think your injectors are spot on as most injectors are controllable up to 92-3%, so you do have some headroom available.

I think the speculation comes back to the fuel pump. I measured flow at the outlet of the pump. There will be some flow/pressure loss as fuel travels through the system, but shouldn't be too significant. The question I have is if the fuel pump assembly itself has any kind of internal restriction that may decrease the numbers I was measuring.

Are you using the complete LS3 fuel pump assembly, with the bucket and fuel pressure regulator?

Also, if you'd like, you can check out the fuel system calculators available on our site > http://www.deatschwerks.com/resources/fuel-calculators/

Low,

Thanks for coming in. Here's a flow table performed by RC Engineering for the LS3 pump at the outlet of the module. Bryan is running the complete module in the car. A bit of re-scaling is needed to adjust the BSFC accordingly.

http://imageshack.us/a/img832/8407/xchj.jpg

At the same time ZL1 and CTS-V fuel modules were also flow tested. The results for the CTS-V are very similar to those obtained by Kinsler for a CTS-V fuel module, so the flow numbers should be good.

One part of the equation that may be missing in yours is the fuel draw required to drive the venturi pumps in the fuel module. If you are taking the flow readings from the outlet of the pump instead of the hat outlet then your flowrates will be higher. Did you remove the pump from the manifold housing for testing?

I was looking at your DW300 last week for a potential project. It's a heck of a pump. Very efficient and nicely packaged.

I had to hook the venturi part of the bucket to the LS3 pump to get a good flow reading out of the pump. With the venturi supply line disconnected and flowing free into the tank, it only supplied 150LPH @ 40psi.

So Bryan, what was the result?

The car is back together and running, and everything is working so far. I drove it last weekend and thought I was capturing data to confirm that the pressure was staying stable, but unfortunately I did something wrong and lost the data. I'll do it again this weekend.

I've noticed that it hold a precise 58 psi at idle, and when I blip the throttle, I see it drops to about 56 and stays there. Again, I need to drive and capture some data to get more exact numbers...

I'll update more soon. I just moved this week, so TOW has been sitting while everything else has been dealt with! :\