98xj 5.3 swap (the poverty method).

[John D]

That said, if you're just looking for an upgrade due to a failing/failed 4.0L, have you considered a crate 4.6L stroker? Same base HP/Torque, roughly the same price (though lots of guys spend $10k on a 5.3 swap somehow... 4.6L crate is under $3k), and takes an afternoon to install.
I waffled between the 4.6L and the 5.3L for years. I went with the 5.3L mostly because I dig a project and the knowledge that comes with such an undertaking.

You should probably wait to make this statement until you get yours in working order. You are obviously and freely admitted by yourself doing things wrong/twice/three-times... When you get it straightened out and correct, I don't think you will be sorry for your decision to go with a 5.3 over a 4.6.

Many of your posts start with or are answered by, "in trying to just get this on the road I did 'X' halfway or different than I intend to and I will go back to fix it later." I also did this a lot on my swap and I can say from experience that in the end I am very happy. Along the way I spent a lot more money and time than I could/should have by not doing it right the first time too. My advice to you is to step back, slow down and work thru the issues completely and systematically. You will get there and it will be worth it.

 

[blistovmhz]

You should probably wait to make this statement until you get yours in working order. You are obviously and freely admitted by yourself doing things wrong/twice/three-times... When you get it straightened out and correct, I don't think you will be sorry for your decision to go with a 5.3 over a 4.6.

Many of your posts start with or are answered by, "in trying to just get this on the road I did 'X' halfway or different than I intend to and I will go back to fix it later." I also did this a lot on my swap and I can say from experience that in the end I am very happy. Along the way I spent a lot more money and time than I could/should have by not doing it right the first time too. My advice to you is to step back, slow down and work thru the issues completely and systematically. You will get there and it will be worth it.

Right now it's pretty near 100% working. I still need to do another longer test drive today, but a quick jaunt out earlier and my LFTF had come down to low single digits (replaced the MAF and fixed fuel pressure), and STFT looks good. The only things I messed up thus far were the fuel pump output hose (not clamped tight enough), and maybe the torque converter install (first bench fit).
Yes, I could spend a tonne more time to make things pretty, but that just doesn't concern me at all. I'll eventually build a new harness, center console, and gauge cluster black box (unless I can get OBD working reliably and with at least 20-30 updates/second).
The only stuff left to do is front drive shaft and transfer case linkage. I may also have to look at the transmission about a possible leak. I don't think I've skipped over anything that needed more attention.

I am very glad I went 5.3. I've already got the same power as a well built and tuned 4.6 stroker, for about the same price, but with better fuel economy and heaps more potential. Once I've got some miles on it and I'm positive everything is working properly (for stock), I'll probably drop a cam in it and get it to a tuner (or just learn to do it myself),
The 4.6 is (from what I've heard) a very nice drop in option, but it's ancient tech and really no aftermarket support to speak of. Having the 5.3, I know I'll be able to get anything I want relatively cheap and in short order. Definitely worth the extra effort.

I don't however, see this as an even remotely viable option for most people. I've got no experience with engine swaps and my fabrication skills are ... ugly, but I know how to research (sorta what I do for a living), and doing an SBC takes either loads of money, loads of research, and/or loads of time. I could've spent a lot more on the swap than I have, but I wanted to see how much I could do myself, and get the benefit of knowing precisely how everything works.

 

[NorIdaWheeler]

You should probably wait to make this statement until you get yours in working order. You are obviously and freely admitted by yourself doing things wrong/twice/three-times... When you get it straightened out and correct, I don't think you will be sorry for your decision to go with a 5.3 over a 4.6.

Many of your posts start with or are answered by, "in trying to just get this on the road I did 'X' halfway or different than I intend to and I will go back to fix it later." I also did this a lot on my swap and I can say from experience that in the end I am very happy. Along the way I spent a lot more money and time than I could/should have by not doing it right the first time too. My advice to you is to step back, slow down and work thru the issues completely and systematically. You will get there and it will be worth it.

This. I spent a lot of time on the small things on my swap and a lot of money in places i felt deserved extra attention. I no doubt could have done my swap for about 2/3 the money i did but i am glad i went about it how i did. I haven't had to redo or go back and give more attention to anything.
I spent lots of time and money in places i consider to be potential issues or safety hazards such as wiring, fuel fittings/hose or anything else that could be a fir hazard. I think the extra time shows in the finished product.

You have done a good job on keeping yourself on a budget! but in my mind with a smaller $$ budget you should allow for a much larger time budget to compensate.

 

[blistovmhz]

This. I spent a lot of time on the small things on my swap and a lot of money in places i felt deserved extra attention. I no doubt could have done my swap for about 2/3 the money i did but i am glad i went about it how i did. I haven't had to redo or go back and give more attention to anything.
I spent lots of time and money in places i consider to be potential issues or safety hazards such as wiring, fuel fittings/hose or anything else that could be a fir hazard. I think the extra time shows in the finished product.

You have done a good job on keeping yourself on a budget! but in my mind with a smaller $$ budget you should allow for a much larger time budget to compensate.

You both may be correct. The budget for this build wasn't at all out of necessity though. I did consider (and could have easily afforded) a crate motor and conversion harness, and off the shelf swap components. But, where's the fun in that? For me, the fun is as much about the project itself as it is the end result. It was more about expanding my knowledge and skill set than having a V8 XJ at the end of the day. The end result is really just icing for me. If I end up spending more to do it my way, I'll have no regrets on how I went about this as I can't imagine any way where I'd have learned more, or got to hone my fab skills further.

While the Jeep is back on the road, I know I'm nowhere near done. There are TONNES of little things that need attention (ie: gaugues, center console, scoop or lowering the engine profile, ladder bars or 4-link, t-case linkage... ). I just didn't want to invest a huge amount of time (or money) into any of this stuff until I knew the power plant was going to work. Same reason I didn't rework the entire harness before dropping it all into the Jeep. If I'd bought a crate motor, or at least something I'd seen run and had an opportunity to test before purchase, I'd have spent more time on the little stuff (ie: harness rework, tear down engine for cleaning/tune up/paint, maybe an ls6 intake and low profile bracketry). The point I'm at now is just the point where I'm confident everything will work, given the time and attention required.

Next on the list is the OBD, because without it, I'm running blind. I'm asking around locally to see if anyone has a known (very) good OBD reader I can test with. I have 3 readers, but I can't vouch for any of their refresh rates or reliability, nor can I find any hard data on the ECM's PID update capability.
Does anyone happen to know how many updates per second the ECM can handle?
My roomies truck (97 1500 4.3) seems to consistently show 20-40 updates/second, but the same reader in my 5.3 sometimes gives me 20/s, but usually 5-10/s. I swapped out the entire serial wire all the way to the ECM, which seems to have beefed up the reliability of the connection, but the refresh rate still usually sucks. Makes it impossible to do any accurate diagnoses of stuff like O2's as I can't see the true waveform. In the 4.3, 40 updates/second is more than enough to accurately gauge O2 status, but in the 5.3, I usually don't see >5 or 10, which just isn't enough resolution.

What could be causing the slow updates? Sometimes I connect and the tach lag is 1-2 seconds, while others it's as close to real time as I can perceive. All I can think is I'm still somehow getting a poor connection with the ECU, and all the OBD applications auto-negotiate the baud rate (silently) until a stable connection is established.
Hell, anyone know the serial commands required to directly interface (like, with a laptop via some serial terminal)? I know lots of guys do it (you have to to write the software), but no one seems to have documented any of it.

 

[yossarian19]

Don't underestimate the value of pro-level hardware and software in OBD readers.

 

[Hoooper]

sounds like you are trying to run too many gauges at the same time. The more information you're asking for over the obd line, the less resolution you get. Does it speed up if you decrease the number of gauges?

 

[blistovmhz]

sounds like you are trying to run too many gauges at the same time. The more information you're asking for over the obd line, the less resolution you get. Does it speed up if you decrease the number of gauges?

No. Like I said, I plug it into my buddies 97 1500 and I get consistent 25hz. Then I drag the same phone and OBD reader over to the Jeep, and get 4-5hz and occasionally up to 10 or 15.

Incidentally, I couldn't find anyone with the same problem but my roomie happened to mention that his wideband O2's were apparently real flaky until he wired their power directly to the same wire feeding the ECM 12V+. So, I gave that a shot and sure as shit it solved the problem completely. Now getting 25-28hz consistently (which I believe is probably the limit of the reader) and connects almost instantly. So that problems outta the way.

Next, I gotta look at the TCC lock-up. I may not have gone fast enough for long enough for it to engage yet, but I feel like I should've seen the TCC slip RPM drop to 0 at least once or twice. 50mph is TCC lockup right?

 

[blistovmhz]

Aaaand I've finally had my holy shit moment. Took it out for a spin to test the gauge cluster and get some fuel. Roads are a bit wet. I'm on very well siped 35x12.5" KM2's. Cruising at 60km/h, give it about 75% throttle and ass end is gone. For some reason the trans is now responding how I expect it to (maybe still learning?). It's shifting much later, and down shifting faster when I need. Steve seriously thinks I might beat his built 5.0L Foxbody on a strip now, given I have the traction advantage.
So, yes, it's fast, and yes, averaging 22mpg combination easy street driving and full drag mode. Still haven't hit highway speeds yet.

 

[NorIdaWheeler]

No. Like I said, I plug it into my buddies 97 1500 and I get consistent 25hz. Then I drag the same phone and OBD reader over to the Jeep, and get 4-5hz and occasionally up to 10 or 15.

Incidentally, I couldn't find anyone with the same problem but my roomie happened to mention that his wideband O2's were apparently real flaky until he wired their power directly to the same wire feeding the ECM 12V+. So, I gave that a shot and sure as shit it solved the problem completely. Now getting 25-28hz consistently (which I believe is probably the limit of the reader) and connects almost instantly. So that problems outta the way.

Im gonna have to try this with mine! good info.

 

[yossarian19]

(misses his old XJ more than ever)
Glad you are enjoying it.

 

[blistovmhz]

Okay, so today I'm going to try to track down another tablet for the cluster, and maybe work on the fuel sender and oil pressure.

1. Jeep fuel sender is 20-240ohms, two wire. GM fuel sender is 40-240ohms, and I assume it's also 2 wire, but there is only one pin on the ECM for fuel level input. Given it's just a resistance measurement, how does the GM ECM pull fuel level? Hooking one end to chassis ground doesn't seem right, as it would interfere with the measurement, but as far as I can tell, the GM fuel sender is also two wire.
2. Still don't know if I have oil pressure . I'm pretty confident that the GM cluster pulled oil pressure from the ECM, as the ECM has an input for the sensor, and the stock cluster is run on a one wire serial bus. I've looked all over and still can't find a PID that works.

Both the fuel level and oil pressure I could just run to a little black box and wire up some LED bars, but I'd really prefer to keep everything in the OBD reader. Any idea's on either of these?

Next up is performance tuning. Down in the US, I'm sure you can go to the grocery store for a $100 tune, but up here there are like 3 or 4 shops in a city of 3million people, and all of them charge $150/h, plus $150 for an HPtuners license, and they estimate a minimum of 4-5 hours on the dyno for a basic tune.$700-$1000 for a basic tune is, like everything else in Canada, insane. I could buy HPtuners myself (almost twice), and have enough licenses to do 4 vehicles for the the price of a single basic tune here.
So, who's done a tune? What is needed? Should I just buy a tuning suite and learn it myself right off the bat (I plan to do this either way, but I'd like to have a good starting point).
Is a basic tune realistic without a dyno (maybe a good long stretch of road?)
I've never done any ECM tuning on any vehicle other than dabbling a bit with Hondata.

 

[NorIdaWheeler]

Okay, so today I'm going to try to track down another tablet for the cluster, and maybe work on the fuel sender and oil pressure.

1. Jeep fuel sender is 20-240ohms, two wire. GM fuel sender is 40-240ohms, and I assume it's also 2 wire, but there is only one pin on the ECM for fuel level input. Given it's just a resistance measurement, how does the GM ECM pull fuel level? Hooking one end to chassis ground doesn't seem right, as it would interfere with the measurement, but as far as I can tell, the GM fuel sender is also two wire.
2. Still don't know if I have oil pressure . I'm pretty confident that the GM cluster pulled oil pressure from the ECM, as the ECM has an input for the sensor, and the stock cluster is run on a one wire serial bus. I've looked all over and still can't find a PID that works.

Both the fuel level and oil pressure I could just run to a little black box and wire up some LED bars, but I'd really prefer to keep everything in the OBD reader. Any idea's on either of these?

Next up is performance tuning. Down in the US, I'm sure you can go to the grocery store for a $100 tune, but up here there are like 3 or 4 shops in a city of 3million people, and all of them charge $150/h, plus $150 for an HPtuners license, and they estimate a minimum of 4-5 hours on the dyno for a basic tune.$700-$1000 for a basic tune is, like everything else in Canada, insane. I could buy HPtuners myself (almost twice), and have enough licenses to do 4 vehicles for the the price of a single basic tune here.
So, who's done a tune? What is needed? Should I just buy a tuning suite and learn it myself right off the bat (I plan to do this either way, but I'd like to have a good starting point).
Is a basic tune realistic without a dyno (maybe a good long stretch of road?)
I've never done any ECM tuning on any vehicle other than dabbling a bit with Hondata.

I found a pid online and got oil pressure to work on my 6.0 in my 2004 Chevy but i have never gotten it to work on my 2002 5.3 in the jeep.
I have been told that the signal is more of a dummy than an accurate reading on some of the ecus. I havent spent much time with it though


If you are going to pay for a tune the first person I would call is Wayne at 150tunes.com He tunes tons of ultra 4 and similar rigs. $150 and you get tunes for life. He is about 30 minutes from me and highly regarded in the offroad industry.

 

[blistovmhz]

I found a pid online and got oil pressure to work on my 6.0 in my 2004 Chevy but i have never gotten it to work on my 2002 5.3 in the jeep.
I have been told that the signal is more of a dummy than an accurate reading on some of the ecus. I havent spent much time with it though


If you are going to pay for a tune the first person I would call is Wayne at 150tunes.com He tunes tons of ultra 4 and similar rigs. $150 and you get tunes for life. He is about 30 minutes from me and highly regarded in the offroad industry.

You're ... where? Somewhere in the central states ?
I'm in Vancouver, BC.

 

[NorIdaWheeler]

You're ... where? Somewhere in the central states ?
I'm in Vancouver, BC.

Im in North Idaho. Just south of you. Im only a couple hours from Seattle, Wa.

 

[Hoooper]

Next up is performance tuning. Down in the US, I'm sure you can go to the grocery store for a $100 tune, but up here there are like 3 or 4 shops in a city of 3million people, and all of them charge $150/h, plus $150 for an HPtuners license, and they estimate a minimum of 4-5 hours on the dyno for a basic tune.$700-$1000 for a basic tune is, like everything else in Canada, insane. I could buy HPtuners myself (almost twice), and have enough licenses to do 4 vehicles for the the price of a single basic tune here.
So, who's done a tune? What is needed? Should I just buy a tuning suite and learn it myself right off the bat (I plan to do this either way, but I'd like to have a good starting point).
Is a basic tune realistic without a dyno (maybe a good long stretch of road?)
I've never done any ECM tuning on any vehicle other than dabbling a bit with Hondata.

Dont pay someone $150 for HPtuners credits, its only $100 worth of credits for the computer you have, so they are charging you an extra $50 for purchasing the credits which if they were a quality shop they should either have an unlimited license on your model already, or have a bulk of credits ready to convert for tuning.

If youre going to pay that much for a tune you might as well buy your own new HPtuners and learn to do it yourself, a new one is "only" $500. You can buy used also, I got mine for I think $300 with all the credits still on it. Then you also have the benefit to make any adjustments you might need to make in the future like if you cam it, do heads, etc.

 

[NorIdaWheeler]

Dont pay someone $150 for HPtuners credits, its only $100 worth of credits for the computer you have, so they are charging you an extra $50 for purchasing the credits which if they were a quality shop they should either have an unlimited license on your model already, or have a bulk of credits ready to convert for tuning.

If youre going to pay that much for a tune you might as well buy your own new HPtuners and learn to do it yourself, a new one is "only" $500. You can buy used also, I got mine for I think $300 with all the credits still on it. Then you also have the benefit to make any adjustments you might need to make in the future like if you cam it, do heads, etc.

I agree to an extent but i enjoy the piece of mind of a professional doing my tune.

and like i said the 150 gets you tunes for life. so making changes for cam, heads etc is a nill point. any changes you make you get a new tune for free.

The op seems like a very diy kinda guy though so im sure he would enjoy doing the tune himself. I guess i just dont trust myself to do it. lol

 

[blistovmhz]

I agree to an extent but i enjoy the piece of mind of a professional doing my tune.

and like i said the 150 gets you tunes for life. so making changes for cam, heads etc is a nill point. any changes you make you get a new tune for free.

The op seems like a very diy kinda guy though so im sure he would enjoy doing the tune himself. I guess i just dont trust myself to do it. lol

For $150, I'd absolutely rather have someone do it right for me the first time, as I have no idea what I'm doing . I'd still end up picking up a tuner, but I know so little about it that I don't even know where to begin at this point. I really hate the way most of the tuners are setup (proprietary stupid handhelds that are just really shitty android handsets instead of just letting you use your own device, confusing/complicated licensing schemes, no documenation until you've purchased the product, etc). I'm an open-source guy. I vehemently refuse to play along however some company wants. If they can't make their product make sense, and let me choose how I want to use their product, I won't deal with them.
That said, SCT does make a tuner that just connects to your Android/iOS device. I've no idea how involved it'll let you get, but I'm looking deeper into it now.

Tuning makes me angry. There is absolutely nothing complicated about changing parameters on the ECM, nor is there anything complicated about the interface (ie: you really shouldn't need a product other than your own laptop, and a serial interface to the ECM). My main client is an engineering company who manufacture stuff like this all day, so I know damned well how simple it SHOULD be to communicate with the ECM. *sigh*.

Took Jeep for another test drive (the woman managed to overheat her car and got herself stranded). Finally got up to highway speed. 20-23mpg seems pretty consistent, but my LFTF did come up a bit from ~0% to 2.3%+ bank 1 and 3.2%+ bank 2. Suspect a very small exhaust leak on bank 2.

 

[blistovmhz]

Trans leak was just the pan. Tightened up a bit and should be good. Will drop the pan later, do a flush, replace filter, and new gasket.

Got lazy and didn't want to attack the dash install tonight, so instead, holographic gauge cluster. I kinda like it a lot. I may stick with it. A few minor tweaks to get the inside reflection up 10x higher than it is, and then I can drop the brightness and you won't see the second reflection from the back of the windshield.

Just need to tweak the layout a bit to pull all the big stuff to the bottom left or mid right (out of field of vision, or at least only in front of the scoop).

Shortened my front DS. ****ed up a bit and busted a tab off the double cardan side so it wobbles a bit around 80-85km/h, but easy enough to fix with a dremel.

Still getting an intermittent open on bank one coil pack, so I'm going to just tear the entire connector apart and install new wires back to the harness trunk. Is a bit scary when you're screaming up a steep hill, in traffic, and all of a sudden you lose 75% of your power. I don't trust automatics cause I don't want to trust brakes

 

[NorIdaWheeler]

The HUD works pretty good in dim light but in direct sunlight it was basically invisible with my tablet. it is a cool feature of torque though.

 

[blistovmhz]

The HUD works pretty good in dim light but in direct sunlight it was basically invisible with my tablet. it is a cool feature of torque though.

Just need a brighter screen . I tested mine at around 7pm and it was still pretty clear. Will have to see again today (if we ever get proper sunlight here in Vancouver).
I may build it into the dash.

Starting to wonder if my oil pressure sensor is just broke. I'm sure I saw OP work at least for a few minutes once.