This got kind of long!
I found a number of problems in the fuel system which took many months to work out. I know - I'm slow! I got an EBL over a year ago and only started to get the VE tables to where I'm happy with them a month ago. I'll just go over everything I can think of.
First, I had what I thought to be a "high flow" fuel filter between the tank and the external fuel pump. After some research I figured out that it was a 10 micron filter. I replaced it with a 100+ micron filter and put the 10 between pump and throttle body.
Next, the EBL showed approx 1-2 volt difference in voltage between the "pump" and "bat". This was basically telling me that my wiring had some problems. Turned out I had the ECM and all loads, injectors, heated O2 and so on coming through a single 16 GA wire (might have been 18) and that was being fed in a round about way through a rat nest of other wiring. I spent a few months rewiring it all to be sure the injectors could get all the current they needed. To be sure the fuel pressure was steady, I ran a new 14 GA wire to the fuel pump to make sure it had full voltage. Oh, and ran an extra 14GA in case of adding a second pump and surge tank in the future.
After that I started on the throttle body. I had plumbed in an aeromotive 13301 regulator and after looking at flow tables, went with the GSL393 to get the higher pressure and flow compared with the GSL395. When that still didn't help, I removed the stock fuel pressure regulator and replaced it with a block off plate. I had it backed off all the way and the 13301 set to about 13.5 PSI, so it should have been overpowered. When that didn't help, I then bored out the throttle body passages to 5/16". The final vertical passage doesn't have enough material so I could only open it up to 1/4". That's the basic problem as near as I can tell and as others said above.
The rest of this is more details on some tests I ran. Hopefully it adds something to the TGO knowledge base.
I have two autometer 15 PSI fuel pressure transducers. I plumbed in one before the TBI unit on the 3/8 inlet line, and one after it on the 13301. Also on the TBI tower, a side effect of boring out the passages to 5/16" is it allowed me to tap it for 1/4" NPT and add a transducer there too. So I had three points to read pressure, one before the TB, one at the FPR and one just after the injectors and before the exit restriction, essentially where the original FPR was and now is a block-off plate.
The pressure transducers are logged by the EBL, and running around town data logging pressures didn't work out well because of what I think is noise on the transducer signals. So I ran a bunch of tests with the motor off, measuring the pressure drop at each point for 3 different fuel flows. I hooked up a jumper wire to each injector so when grounded they would be static. That gave:
0% DC -- fuel pump jumpered on, injectors closed
50% DC -- fuel pump on, 1 injector grounded
100% DC -- fuel pump on, both injectors grounded
(This was a little tricky to catch the fuel in the TB without flooding out the engine - it's amazing how much fuel an injector puts out at 100% DC!)
As a side note, I found that each injector needed a 10A fuse.
First I calibrated the transducers. The aeromotive FPR holds pressure for a long time - many minutes. First key on to pressurize the system then key off for no flow. Then measure the transducer outputs. For what it's worth they differed by 10% - I bought one of them used and the other was new, that's the only thing I could think of to explain the difference.
With IGN on, motor off, I'd start data logging then ground the injectors in sequence. Key on engine off for 5-10 sec, ground one injector for about 1-2 second, then ground the other for 1-2 second. That gave me the three readings in one datalog. Looking back at the datalog I could see where the pressure dropped from 0% DC to 50% and 100% and by how much.
Next I did the same experiment with the original lower flowing GSL395 pump. Then charted everything with excel.
My conclusion was that the design of the TBI unit with FPR just after the injectors and before the restrictions in the TB worked better than the external FPR. I removed the 13301 and went back to stock FPR, with the pressure screw cranked up to it's max (about 13.4 PSI with fuel pump on, motor off). The aeromotive was cool, but the EBL makes it so easy to adjust the BPW that it's a simple matter to read the fuel pressure and adjust the ECM rather than the other way around. Maybe it's backwards but that's my logic. I also went back to the stock TBI lower flowing GSL-395 pump. This pump flows less at idle and I convinced myself that it was fine at 100% DC.
As you can tell by now, my first venture into tuning was to start with a stock motor and learn from there. In the future, if I ever build a nice motor that needs a lot more fuel my plan is to go back to the external areomotive 13301 regulator and the drilled out TB pod. I'll run the fuel input into both stock fuel connections - the inlet and outlet - like it's dead-headed, only I'll add a fitting plumbed into the block-off plate with a large diameter return back to the FPR via the hole in the bottom of the TBI unit. That would allow it to flow the full amount of output from a high flow fuel pump at idle. Otherwise it seems to me that it's hard to tune accurately for a VRFPR with the added variable of pressure drop in the TB passages. Of course, maybe the tuning doesn't have to be that close, I don't have a feel for that yet. With the current setup, basically stock, the BLMs sit on 128, and maybe hit 126 or 130 once in a while. With the external 13301 they were all over the place.
I took a few photos of the setup and have the flows charted with excel, if anyone is interested I can post them up.
? Shouldn't the FP hold steady? 
