Jump to content
Ultimate Subaru Message Board

GeneralDisorder

Members
  • Posts

    23391
  • Joined

  • Last visited

  • Days Won

    435

Everything posted by GeneralDisorder

  1. Well - you can use an aftermarket coil but it risks damage to the ignition amp in the distributor (perhaps what has happened to the 2WD unit?). But for the 4WD distributor the ignition module is only about $25 on Amazon. Makes them pretty cost effective to repair and keep a spare replacement around. Only need a couple flat-head screwdrivers or small pry-bars and a #2 phillips to replace the module. GD
  2. I probably have some heads, yes. They would be 2.2 heads but they are interchangleable with the 1.8. What's wrong with the heads you have? Usually pretty cost-effective to rebuild them.

     

    Rick

  3. They have proven to be reliable well past 300k miles. Don't touch it. Subaru's timing chain setup is amazingly reliable - in fact they have adopted it for the new FB 4 cylinder engines that came out in '11. It's also extremely expensive to repair or replace should it have to be. There's something like 117 bolts that hold on the front cover. The biggest problem the H6's have is the AC belt tensioner bearing :-p. GD
  4. You also need the matching coil. For a 4WD that means you need a matching Hitachi coil in place of your ND coil. GD
  5. Limited in what way? The tuning limitation is what this thread is talking about overcoming. If the EJ20G ECU can run the 22T so easily it's because they have essentially the same hardware minus the 4 channel ignition. That means that any "limitations" are software based and that's the whole point of hacking the code - to overcome the software limitations. When/if we have complete software control of the stock ECU it BECOMES a standalone with most of the same features and abilities. Besides all that - I don't really care what other people want to do or what they want to use. I'm more interested in the process of reverse engineering the code on an ECU.... any ECU really.... as a learning experience. This just happens to be the one I own and daily drive so it's accesible to me and I don't have to pointlessly rewire my car. I'm a former full-time software engineer and I enjoy the experience and challenge. GD
  6. They seem to be laying around junk yard shelves all over the nation based on car-part.com. There's also the legacycentral bbs. They can be found and they have a very low failure rate. OBD-I vs. II is not really that helpful from a tuning perspective - once you have control over the maps and RTD streams there aren't any real differences - it's really just a communications protocal preference. There is nothing wrong with the SSM interface. GD
  7. Again - that's all well and good if all you want is an off-the-shelf tune on a chip and you don't care about ever changing anything or tinkering. Personally I like to play with things - and the 22T ECU is perfectly cabable of doing what needs to be done - not to mention it's a lot more accesible - easier to source replacements. There are a lot of folks that would like to not have to source a 20G ECU and not have to mess with their wiring, and not have to rely on someone who sells tuner chips. GD
  8. Very cool. When I get a spare minute I'm going to pull my ECU down and get some pics.... GD
  9. Just the part or installation as well? The radiator will run about $100 to $150 depending on your sources. Should also replace the hoses - get those from the dealer. Installation is typically about one hour. I do a fair number of radiators - usually it's about $300 with radiator, hoses, and labor. GD
  10. Certainly seems like the relay is bad then. Possibly rusted/corroded inside. Cut it open and look. It's an informative experience to open a relay and see it's internals. Then it will be clear to you how that particular unit functions. GD
  11. Compression test is a waste of time. Pulley comes loose, key and keyway are mangled or sheared, timing is affected, car won't start. Bring it to me. I'll take care of it for you. A compression tester isn't the tool you need - a welder and the skill to use it is. GD
  12. What do you have at the yellow wires going into the ignition relay? GD
  13. Time for a complete timing belt/WP job. The key has sheared off and the crank sprocket has slipped on the crank. Depending on the year it could also cause internal engine damage. If you need help give me a call: Superior Soobie & Import, 503-880-4084. I'm in Milwaukie. I've fixed a number of these without replacing the crank or the engine as a lot of shops will tell you needs to be done. GD
  14. Surge tank should always be full so you shouldn't have to do anything but run both pumps - add a second relay and run it from the same signal as the existing fuel pump relay. The only time you would ever have to worry about it is when you need to pre-fill the surge tank - like the first time you start it or if the system has been purged of fuel. For that I would add a momentary switch under the dash somewhere to cycle the delivery pump as long as it's held down. All that switch would do is ground the delivery pump relay when you have the key on but are not cranking it. GD
  15. Electronics can fail from age. Especially capacitors like to start leaking and transistors will occasionally burn out from heat cycling. But I would guess that you are fighting a resistance battle with your grounds or power supplies. Checking the voltage and the resistance per the FSM is not sufficient to tell you if the wiring will carry a 10 amp draw without voltage drop. You really need to perform a voltage drop test with a load in the circuit that approximates the load of the EGR solenoid. You may have a corroded ground connection or a bad connection at your fusible link that supplies power to the ECU.... for example. These come in handy for what you are about to test: http://www.esitest.com/180.html GD
  16. So - at the plug to the ignition relay you have 12v at the yellow wires but when the relay is energized you have 5v coming out on the yellow/red? Bad ignition relay. GD
  17. That's the wrong way to do it. The ignition relay is there for a reason. You need to find the source of the voltage drop, not remove components and wire around them. Hard to say - probably missing a ground source or power source. No. The pump will continue to run in an accident scenario and could cause a fire or fuel spill. I've stripped more harnesses than I can count and it's never been wrong that I've seen. Completely uneccesary and a waste of valueable time and resources. The ECU already handles the fuel pump shutdown by cutting out the fuel pump relay when there is no ignition pulse (tach signal) present. The wiring, hardware, and software is already in place to support the "safety" you desire so adding additional components and wiring is just wasted effort. GD
  18. For sheet metal you really need thin wire and shielding gas. The flux will make it impossible to do short tack welds as the slag needs to be removed before striking the arc to prevent slag inclusion in your welds. You don't actually have a "MIG" welder - common misconception. You have a "wire feed" welder, or FCAW (Flux Core Arc Welding). MIG stands for Metal Inert Gas welding and if the machine isn't capable of using gas then it's technically not a MIG welder at all. Without a true MIG setup, .023 wire, and sheilding gas (preferably 100% CO2 as it runs colder) you are better off with pop rivets and aluminium sheet. FCAW for automotive sheet metal work is really difficult even for those of us that have some welding skills. GD
  19. 97 and 98 have adjustable lash and require adjustment ever 105k miles. 90 to 96 are hydraulic and no adjustment is possible or required. The '95 and '96 are cross-over years with hydrualic lifters and roller rockers. GD
  20. I don't want a chip with someone's propietary code on it - I want to be able to modify the parameters myself as well as log the real time data streams. GD
  21. This is really excellent stuff that I hadn't seen before - seems like every few years when I look into this subject more and more information is availible. I'll pull out my ECU and take some pictures to post here. I would like to swap over to EJ25D heads and manifold as I have tons of those components - I also have a TD05-16G and a Forester XT intercooler - I should be able to use most of the EJ22T components on the 25D manifold but the injectors will definitely change and of course there would be little point to all that work if there wasn't a bigger turbo and more boost pressure to be had as well. All that requires tuning that has, in the past, required a standalone to do properly. It just seems like such a waste when the stock ECU could handle what I need with just a few tweaks to it's ROM. GD
  22. All dual exhaust port engines will have non-roller rockers except the '95 engines. This engine can be identified by it being OBD-II vs. OBD-I - IE: Three wiring harness connectors at the upper right of the bell-housing instead of two. As to which is best.... there's no clear winner. Both seem to have the same reliability and longevity. The roller rocker components are harder and should show less wear in the long run I suppose. It's not till '97 that they went to solid lifters which will decrease drag on the valve train and should allow a slight performance and effeciency enhancement. The rockers and camshafts are different. You shouldn't use a non-roller cam with roller rockers as the roller cams are harder to keep the rollers from chewing them up. No. They are all the same with respect to reliability. You can if you know what you are looking for. GD
×
×
  • Create New...