oldred95

Finally playing with the MAF again and its running, that wasn't any big deal but its still not shifting (4L60E). Between Fast355, dimented24x7 and myself we found a few little things that weren't set right and we fixed that but it still doesn't want to shift. If I'm not mistaken the only thing thats done is the VSS is hooked directly to the PCM at pins F4 and F5 and then the PCM is told to look at pins F4 and F5 for its speed signal and not at pin F12 like it does in stock form. Seems simple enough to me. Well I've tried various things, all the ratios for converting TOS to MPH should be correct and so on. My speedometer works still so it is obviously seeing vehicle speed but in the datalog my vehicle speed is always zero mph. Also I took a log with datamasters and it shows the active gear to always be 3rd gear no matter if its in park or any gear position. Right now since the PCM isn't seeing vehicle speed its defaulting the transmission to limp mode and I've only got reverse, 1st and 2nd which have to be manually shifted. All I have to do is change a chip and repin one wire and I'll be back up and running in speed density but I was wanting to get this going yet this weekend. Any ideas on what I'm not doing right and why the PCM won't see vehicle speed?

Fast355

iTrader: (2)

Does it set a transmission speed sensor code?

BTW, I am logging onto AIM, send the log.

oldred95

The SES light was on but I have no idea what code it was for.

dimented24x7

iTrader: (2)

As I said earlier, use your stock code with the previously mentioned option bit set to try out the alternate transmission speed input so there are fewer unkowns. On the bench the input works fine, and also works well with generic magnetic pickup speed sensors, but the buffer box is an unkown and Im not sure if it puts a bias on the signal. The input will trigger with as little as +/- 150 mV, but if it fails to cross 0V, the input wont trigger, and youll have no forward gears (computer thinks the truck is stationary). One way to test this is to take a DVM and measure the voltage between the VSS low and high. If there is some sort of a bias there, then the PCM may not be triggering. If it still fails to work, try unplugging the buffer box to take it out of the equation.

Fast355

iTrader: (2)

We are trying something a little different. I read a post where Junkcluter had traced the circuit back on his PCM. The F4/F5 circuit was missing capacitors and resistors in his PCM. I have relocated the input for the VSS calculations to the TIS counter and he is trying it out as I type this. We will find out very soon if it will shift or not. I am 100% certain that his PCM will have the F2/F3 input circuitry there, but not on the F4/F5.

Edit, this is the code changes that I have done

;~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
; Determine transmission output shaft speed
;~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
;
LDX #$3027 ;TOS pulse accumulator (Disabled)
LDY #$3010 ;TOS 65,536 Hz free running counter, captured
;on each edge detected (disabled)
;
; SPD SENSOR SOURCE
; 0010 0001
; b0 1 = Allow tps hist buffer every 25 Msec
; b5 1 = Force 2nd Gr if in D2 and not manual
;
; b4, 0 = Output speed from dig. ratio adaptor.
;
LDAA #$10 ; b4
BITA L5D02 ;Opt. byte, test b4
BEQ LB0B2 ;Bra if ==0, speed from dig. ratio adaptor
;
LDX #L3027 ;Alternate TOS PA
LDY #L3011 ;Alternate TOS counter


L3010 EQU $3010 ;MSB, 131 kHz free running TIS counter
L3011 EQU $3011 ;LSB, 131 kHz free running TIS counter

L3027 EQU $3027 ;TIS pulse accumulator

dimented24x7

iTrader: (2)

Wow. Now thats interesting that only some came with it enabled. Even though the TIS PA is only 8 bit, it should still work well for the TOS as its read fast enough that there wont be consecutive overflows, and it will have more resolution. The only things that need to be done other then replacing X counter/PA with Y counter/PA in the code, is to divide teh timer value by 2 as the constants for the TOS calcs are only 16 bit. IOW, the constants will be out of range if you use a 131 kHz free running counter. The following will need to be done:

Add the following to routine E34C:

LDD 6,Y ;Load captured free running counter value
SUBD 1,Y ;Subtract counter val., one delayed, now time between
;calculation interval, which is secs x 131072
LSRD ;/2, now secs x 65536
XGDX ;Swap revs into D, elapsed time into X
FDIV ;[RPM x (elapsed timer/65536) x 8] /
;(elapsed timer/65536) = TOS RPM x 8

Also keep in mind that the TIS PA is only 8 bit and not 16 bit, so youll need to load it into registers A and B (clearing A) and then swap it into register X for the calculations. In addition, the timeout value may need to be changed as the free running counter rolls over after .5 secs. If you need some more help with this, just LMK and Ill take a look at it.

oldred95

So you see what the problem is do you? Its like a told Chris this afternoon, what we have here is failure to communicate.

Tbi-MAX

hey oldred, be sure to tell me how it goes, as i may do the MAF mod after you get yours up and running, as we basically have the same truck. but i remember when my map was all messed up when i went from SBC to BBC.. the truck had no clue when to shift or change gears...

oldred95

Its not a MAP related thing. The PCM is not seeing vehicle speeds. It thinks I'm sitting still in park while I'm actually moving. It does run incredibly smooth though on the MAF for only being ran a few minutes and not really learning what needed adjustments where.

dimented24x7

iTrader: (2)

Thats why I asked that you try the stock code first before the MAF code, so that I could see what part of the system wasn't working. The two j-yard cores (93 V6 and a 95 V8) that I had both have that input enabled (thats how I found it), so the assumption was that all the units would have the input. Evidently, though, only select computers had the input enabled if Junk says that the resistors and caps where missing. Sort of sucks as you cant run a 4L80-E as easily.

oldred95

It was always my understanding that pins F2 and F3 where for the Turbine Input Sensor on the 4L80E and that they would work to read the Vehicle Speed Sensor regaurdless. Pins F4 and F5, if I'm understanding this correctly are the ones without resistors and capacitors. I still don't see why it won't work.

dimented24x7

iTrader: (2)

F2 and F3, assuming those are terminated as well, will work just fine. Just need to account for the fact that the counter runs at a different rate. Sounds like junk and fast are testing that as we speak. Id test it, but it appears that I have earlier computers that have everything terminated and operational. I dont want to say it works fine and have it still be dead when you go do drive it. Mine also have the component #s and types silk screened onto the PCB, which seems to only be on some of the PCMs. It would seem that at some point, GM decided to save a few 10ths of a cent on each unit produced by eliminating the extras. Go figure :/

The only issue that could arise with the 4L80-E is the lack of the turbine input speed (on pins F2 and F3). IIRC, this is used in the apply/release logic for the PWM'd TCC. I would assume that it would work fine just using the TOS to deduce the TIS, but theres no way to know for sure. It could potentially be choppy when it applies or releases.

oldred95

Well it sounds like it will work then. Just going to take some more R & D before it works.

pfk

Once again, you posted about something I was searching for

I took the truck out for a drive after converting to the MAF code and have the same issue as you. I have 2 or 3 or 7427's here so based on what I read here,I figure I might try swapping them in to see if any of them are built different.

I'm subscribed now and I'll post back if any of the other ecms work for me, in the meantime I'll keep searching on tuning the fuel with the MAF

oldred95

Its looking like a PCM swap might be the easiest fix but how do you know if a pcm is going to have the credentials to work with the MAF? Seems like a PCM from a 4L80E vehicle would deffinitely work would it not?

pfk

I wonder if the 4 digit code after the service number is what could identify it.
I have a BJYK, BJYL, BJLA, so I can try those 3 and see what happens (the BJYK is in there right now and not working)

oldred95

I have BJYL and for me it isn't working. I don't know what fast355 has but he somehow made the MAF and 4L60E setup work on his brothers S10 blazer but it was with a different speed sensor and kind of a custom setup but still the principal should apply here. He just can't remember how he did it.

pfk

None of the 3 I have will work by just plugging it it, but it was worth a try.

if there's anything I can do to help by testing,let me know and I'll try it out, I'll be fooling around with the truck pretty much right through the weekend.

pfk

I'm in the truck fooling around as per the rest of the info in this thread and I just noticed something different.

If I do not have the TOS from the buffer to pin F12 hooked up at all, My speedo works along with the datalog reporting vehicle speed but obviously no trans shifting(I'm assuming still working off of the input from buffer at F13) When I "T" off the vss wires directly to F4 and F5, I lose the speedometer and pcm input. I tried this with all 3 of my pcm's and then tried it again on pins F2 and F3 with all 3 pcm's. So when I hook vss directly to pcm, I lose input to F13 from buffer therefore I get no shifting but no speedometer either.

I'm assuming I must be missing a step in the wiring process or something....

dimented24x7

iTrader: (2)

The speedo is run off of the buffer box. No signal from the PCM is needed. Pin F12 was the TOS input, but is now taken up by the MAF. Pins F4 and F5 have a polarity. One is the reference, the other is the signal. The buffer box likely has the same type of input. If the signal from one and reference from another are on the same line, the signal from the VSS will zero out and the speedo/ABS will be kaput.

Oldred, did you ever figure out what that error was?

pfk

You hit the nail on the head with that one, swapped the wires around and speedo is working again.

I added a flag/switch to the xdf for the TOS input, could someone confirm that I have done it correctly?

oldred95

Sure haven't. I'm go back to school this tuesday and I haven't had time to monkey with it. I'll throw it back on the MAF this next weekend since I won't be doing anything anyway and see what the code is.

pfk

With the wiring hooked up as it should be, I went back to the SD bin and checked the above mentioned bit (as per my example)

The vehicle starts off in first gear and won't shift.

When I am using the MAF code, it starts off in third gear. (I did not set the option in the MAF bin as I assumed it would point the software to look at the TOS input for the MAF)

I took a couple of my pcm's apart for the hell of it and they are indeed missing the caps at the F4 and F5 inputs.

Just out of curiosity, is that the only part of the circuit that is missing? and if yes can I just solder a couple of caps to the board and have it work? (I thought that I could maybe steal the caps from F2 and F3 from one of my other pcm's but it would be nice to know if they were the same first)

dimented24x7

iTrader: (2)

I guess I really lucked out with my units. They both came with the input operational. The stuck in first gear with the SD code means that the input is not functional. The stuck in third gear with the MAF code means that the E-trans isnt enabled. Your version may also be an older one with the TOS disabled rather than relocated.
----------
The caps are likely different, but it may not matter. However, the knowledge that theyre there on the TIS means that relocating to F2/F3 like I have it now may be the solution for the e-trans setups.

pfk

Is there a new version available or should I wait until more testing is done?

pfk

I unchecked Flag L400F Bit 0 as per oldred95's email, and it worked like a charm. The truck seems to shift normally through all ranges and drives well.

I was out driving it last night and I'm finally at the point to make adjustments to AE (at least I think so )

Fast355

iTrader: (2)

Sweet, so we have MAF TBI and a 4L60E for all now.

Truck running smoother and more consistant on MAF for you as well?

pfk

most definately. I was having a tough time getting my SD tuning going and had just started to figure it out when I tried the MAF code, had it running better after on MAF in one hour than it was on SD tune after 2 weeks

I couldn't find any info on the calibration of the MAF sensor , I guess I dont know where to look yet, but I used my SD tune to give me a rough starting point and changed it according to the WB after. A couple of days later, it starting running like crap, and I kept trying to make adjustments with horrible results (kept running universally rich) I went back to fuel system, nothing wrong there. On a hunch, checked the air hoses to the TB and found a leak, as well as my bonnet virbrated a little loose.I tightened everything up and it ran better than ever, frequency hits approx 3500 at idle compared the the 4000 that I was idling at originally.

I ended up with this as my flow vs freq table (starting at 12000Hz) using 85mm MAF from 99 5.3 (part#25168491) I did notice that I was running lean last night when I was testing out the tranny side, so I may end up recalibrating if it wasn't something else I changed (I'm still doing lots of learning)

165.47
165.47
165.47
165.47
165.47
165.47
165.47
165.47
161.65
155.61
149.86
144.39
139.18
134.20
129.47
124.90
120.54
116.38
112.37
108.52
104.81
101.25
97.79
94.46
91.23
88.11
85.08
82.14
79.26
76.47
73.77
69.75
66.95
63.55
61.13
58.10
55.39
53.24
51.73
48.30
45.98
44.84
43.97
40.14
35.80
32.99
30.32
28.20
26.91
25.67
24.52
23.30
22.17
20.96
19.61
17.62
16.22
15.16
13.91
13.57
12.34
11.78
11.12
10.49
10.06
8.69
7.98
7.23
6.53
5.90
4.79
4.19
3.65
3.14
2.67
2.25
1.76
1.22
1.02
0.75
0.50
0.39
0.37
0.37
1.37


I think the highest frequency I have hit so far was 10500hz, but that was when I was logging while driving around in 1st gear only.

Fast355

iTrader: (2)

No wonder it is running lean. This is what I had for my heavily moded, NA vortec 350 TBI.

428.46
414.53
400.92
387.53
374.64
361.96
349.59
337.51
325.73
314.25
303.05
292.15
281.53
271.19
261.13
251.34
241.83
232.52
223.52
214.80
206.34
198.21
190.23
182.58
175.22
168.08
161.15
154.41
147.87
141.53
135.30
129.24
123.35
117.63
112.09
106.71
101.51
96.49
91.63
86.94
82.40
78.03
73.80
69.73
65.82
62.05
58.44
54.98
51.66
48.50
45.48
42.62
39.90
37.36
34.95
32.66
30.49
28.41
26.43
24.54
22.70
21.02
19.32
17.70
16.21
14.85
13.60
12.45
11.38
10.38
9.45
8.56
7.73
6.95
6.22
5.54
4.91
4.34
3.80
3.33
2.89
2.49
2.13
1.80
4.00

I was running 68# injectors at 32 PSI, with a GM Marine regulator.

14.12
13.75
13.42
13.09
12.74
12.39
12.01
11.63
11.21

pfk

I wonder how the math would relate the 2 since I am using a fudged number for the injector constant that I needed to keep the ve's in range?

actual Gms/sec should be about 19.9, and I am running it at 10.


Yours appear to be 2x as high as mine are (with the exception of the areas I have not yet touched where yours are almost 3x higher) Since I am running 2x the amount of fuel through the injector that it thinks it is running, could this amount for the large differences in the 2 tables?

I tried hooking up the reference to the vafpr with the actual amounts but my ve table would be way outta range again. I wasn't sure how to handle the multiple entries so I set them all at 10, knowing that my ve's would be in check.

should I go back to the actual amounts and forget about the ve table?

Fast355

iTrader: (2)

The difference in injector flow rates inputed into the PCM would make a HUGE difference in the tables. I would believe that the two are inversely related. That means if you double the fuel flow rate, the MAF table will be cut in half and vice versa. You might take and multiply the higher flow stuff by 1.5 and see where that gets you.

I am a firm believer in not fudging the numbers to the PCM or if I do, the least amount possible. Like when VE tables are getting close to maxed out with the older ECMs, I would decrease the flow constant to lengthen the pulsewidth.