Megasquirt + stock injection
-
- Posts: 23
- Joined: Tue Jan 12, 2010 7:17 pm
Megasquirt + stock injection
I posted something about this on the Samba as well, but wanted to try here due to the specialized FI forum. I'm looking into upping my displacement to 1776 in my '78 engine with stock injection. The system works fine with no major issues and all the parts look to be in good shape (the engine is completely apart at the moment). Somebody suggested that I would have to change the controller to Megasquirt or something similar so it can be adjusted to the changes. How does it work with stock injection, and what type should I get? I'm good with electronics and soldering so I don't mind getting a kit if needed.
Also, do the mechanical parts of the system have enough juice to support an upgrade to 1776?
Thanks
Also, do the mechanical parts of the system have enough juice to support an upgrade to 1776?
Thanks
'78 Super Convertible restoration in progress!
- raygreenwood
- Posts: 11895
- Joined: Wed Jan 22, 2003 12:01 am
Re: Megasquirt + stock injection
Replace everything but the intake manifold and TB with MS. The stock intake while not being perfect should work quite well (as you have found) with a 1776. Do not solder anthing in EFI. Crimp only...never solder. Ray
-
- Posts: 3
- Joined: Sat Apr 10, 2010 10:59 am
Re: Megasquirt + stock injection
Why is this?raygreenwood wrote:Do not solder anthing in EFI. Crimp only...never solder. Ray
-
- Posts: 23
- Joined: Tue Jan 12, 2010 7:17 pm
Re: Megasquirt + stock injection
I'd like to know too. Is this something specific to EFI or some other reason with regards to assembling boards?
'78 Super Convertible restoration in progress!
- Chip Birks
- Posts: 4003
- Joined: Wed Mar 12, 2008 5:59 pm
Re: Megasquirt + stock injection
Ray is talking about the wire harness itself. Not the ECU, solder the ECU to your hearts content, but try to do as much crimping as possible with the harness.
-
- Posts: 3
- Joined: Sat Apr 10, 2010 10:59 am
Re: Megasquirt + stock injection
I know the megasquirt itself is soldered together. I don't understand why a crimped connection is superior to a soldered one in the harness.
-
- Posts: 5
- Joined: Tue Aug 11, 2009 6:26 pm
Re: Megasquirt + stock injection
this is a famous internet debate (well at least on some of the other forums i'm on)
people say soldered connections can break at the joint. I think it has to do with how much stress or movement is seen at that joint. For me, I still solder and heat shrink to add further strength (prevent bending at the joint). 9 out 10 times the solder joint is inside of a plastic connector which to me adds further protection.
people say soldered connections can break at the joint. I think it has to do with how much stress or movement is seen at that joint. For me, I still solder and heat shrink to add further strength (prevent bending at the joint). 9 out 10 times the solder joint is inside of a plastic connector which to me adds further protection.
-
- Posts: 3
- Joined: Sat Apr 10, 2010 10:59 am
Re: Megasquirt + stock injection
That makes sense. The soldered connection to a speaker in the driver's door of my jeep just broke a month ago. Soldering a crimped connection seems like insurance to me though.
Back on topic:
I feel that the stock hardware should do fine on a 1776. Replacing the stock ecu with a programmable one would be necessary to get it running properly though. I would think the stock ecu could run it, just not to the full potential of the increased displacement. I hope others agree with this as it is my plan in the near future as well.
Back on topic:
I feel that the stock hardware should do fine on a 1776. Replacing the stock ecu with a programmable one would be necessary to get it running properly though. I would think the stock ecu could run it, just not to the full potential of the increased displacement. I hope others agree with this as it is my plan in the near future as well.
- Joe Perez
- Posts: 44
- Joined: Thu Dec 04, 2008 2:26 pm
Re: Megasquirt + stock injection
Indeed. Open up a factory harness, and you'll see very few (if any) soldered connections. Folks generally take this as an affirmation of the superiority of crimping for wire-end termination.
There are a couple of factors to consider, however.
One is that from a manufacturer's perspective, criming is highly desirable as it is a massive time-saver. A worker sitting at a bench with a semi-automatic crimping machine using terminals on a reel can bang out hundreds of terminations per hour. You just insert the stripped end of the wire into the machine, hit the foot pedal, and the machine automatically locates the wire, performs a multi-level crimp, cuts the pin off of the reel, ejects it, and advances the next pin into the die. We have a couple of these machines in our shop for doing AMP Mod_IV and Molex 062 terminations, and its simply amazing to watch the folks down there work. By comparison, when our folks do encounter terminations that much be soldered (such as XLR and 1/4 TRS connectors) they are much slower at these, probably by a factor of ten.
Another thing to consider, for the home user, is the quality of the crimp. The machines that I mentioned above do a really phenomenal job, making perfect two-level terminations with proper foldover every time. The downside is that they're expensive. Even the hand tools that we use to do these terminations in the field cost several hundred dollars each. Here's a link to one of the many crimpers that I carry in my toolbox: http://search.digikey.com/scripts/DkSea ... =A99114-ND It costs $570. And I have a half-dozen or so of these for different types of terminals.
And of course, the type of terminals we're dealing with are light-years away from the insulated crimp terminals you usually see at the auto parts store- they have separate levels of termination, where one level makes the conductive contact, and another forms a strain-relief on the insulation. And most folks don't have the correct tool to achieve a good termination even with these. They generally own something that looks like this:
Here's a tool similar to the one I use for insulated barrel connectors (ring terminals, butt-splices, 1/4" quick-disconnects, etc.) This one's only $300: http://www.newark.com/panduit/ct-1550/c ... dp/99F7058 You can get cheaper versions, of course. This one is about the absolute minimum that I'd consider usable for anything resembling critical wiring: http://www.newark.com/molex/64016-0036/ ... dp/52M2114
With all that said, I've got a heck of a lot of soldered connections in my car. In particular, anything that's in a hostile environment (such as ring terminals used underhood) first get crimped with my fancy tool, and then I apply a bit of solder at the very end of the wire, to ensure a robust and corrosion-free electrical contact. (It's important to use enough heat so that the solder flows into the connector- this isn't a place to wimp out. If you have liquid flux in a bottle, that helps, too.)
There are a couple of factors to consider, however.
One is that from a manufacturer's perspective, criming is highly desirable as it is a massive time-saver. A worker sitting at a bench with a semi-automatic crimping machine using terminals on a reel can bang out hundreds of terminations per hour. You just insert the stripped end of the wire into the machine, hit the foot pedal, and the machine automatically locates the wire, performs a multi-level crimp, cuts the pin off of the reel, ejects it, and advances the next pin into the die. We have a couple of these machines in our shop for doing AMP Mod_IV and Molex 062 terminations, and its simply amazing to watch the folks down there work. By comparison, when our folks do encounter terminations that much be soldered (such as XLR and 1/4 TRS connectors) they are much slower at these, probably by a factor of ten.
Another thing to consider, for the home user, is the quality of the crimp. The machines that I mentioned above do a really phenomenal job, making perfect two-level terminations with proper foldover every time. The downside is that they're expensive. Even the hand tools that we use to do these terminations in the field cost several hundred dollars each. Here's a link to one of the many crimpers that I carry in my toolbox: http://search.digikey.com/scripts/DkSea ... =A99114-ND It costs $570. And I have a half-dozen or so of these for different types of terminals.
And of course, the type of terminals we're dealing with are light-years away from the insulated crimp terminals you usually see at the auto parts store- they have separate levels of termination, where one level makes the conductive contact, and another forms a strain-relief on the insulation. And most folks don't have the correct tool to achieve a good termination even with these. They generally own something that looks like this:
Here's a tool similar to the one I use for insulated barrel connectors (ring terminals, butt-splices, 1/4" quick-disconnects, etc.) This one's only $300: http://www.newark.com/panduit/ct-1550/c ... dp/99F7058 You can get cheaper versions, of course. This one is about the absolute minimum that I'd consider usable for anything resembling critical wiring: http://www.newark.com/molex/64016-0036/ ... dp/52M2114
With all that said, I've got a heck of a lot of soldered connections in my car. In particular, anything that's in a hostile environment (such as ring terminals used underhood) first get crimped with my fancy tool, and then I apply a bit of solder at the very end of the wire, to ensure a robust and corrosion-free electrical contact. (It's important to use enough heat so that the solder flows into the connector- this isn't a place to wimp out. If you have liquid flux in a bottle, that helps, too.)
- raygreenwood
- Posts: 11895
- Joined: Wed Jan 22, 2003 12:01 am
Re: Megasquirt + stock injection
The endless debate. There has not been one single soldered connection in any production fuel injection system...ever....for very good reasons. Vibration, heat, expansion, voids etc. etc. etc. etc. etc.
No one cares how well you solder. Soldered connections are inferior in lifespan, resistance etc.
Actually...crimping is about four times MORE expensive than soldering....because of teh cost of the connector alone. The terminals alone are extremely expensive. Factories quit soldering ANY terminals other than hard connections to boards...decades ago..because of reliability issues.
All soldered connections from the late 60's on...were machine soldered. Crimps are machine crimped. This myth that crimping was started as a money saver...is BS.
Its strictly reliability.
Did you know that properly crimped terminals are hermetically sealed? The gap between terminal and wire is non-existant. Can you guarantee that with with soldering? I didn't think so.
Also...soldering creates brittleness in the very thin connector systems of the EFI harnesses. Try heating one to anything over about 350 and see what happens. Been there and done that.
The plating systems and level of spring in the terminals of EFI harnesses are critical. Heat kills that. Why...when you have a nearly perfect connection with crimping...would you think that soldering that terminal...whose plating seperates with heat...and whose crimp loosens with heat becasue it expands the crimp...and generally screws up the crimp connection...would be a safeguard?
Its nice that you can solder and believe in it....but it has no place in fuel injection harnesses.
Do some research please into this please. You can buy a proper crimper...with dies to do every connection within your car just like factory....for between $60 and $100. Ray
No one cares how well you solder. Soldered connections are inferior in lifespan, resistance etc.
Actually...crimping is about four times MORE expensive than soldering....because of teh cost of the connector alone. The terminals alone are extremely expensive. Factories quit soldering ANY terminals other than hard connections to boards...decades ago..because of reliability issues.
All soldered connections from the late 60's on...were machine soldered. Crimps are machine crimped. This myth that crimping was started as a money saver...is BS.
Its strictly reliability.
Did you know that properly crimped terminals are hermetically sealed? The gap between terminal and wire is non-existant. Can you guarantee that with with soldering? I didn't think so.
Also...soldering creates brittleness in the very thin connector systems of the EFI harnesses. Try heating one to anything over about 350 and see what happens. Been there and done that.
The plating systems and level of spring in the terminals of EFI harnesses are critical. Heat kills that. Why...when you have a nearly perfect connection with crimping...would you think that soldering that terminal...whose plating seperates with heat...and whose crimp loosens with heat becasue it expands the crimp...and generally screws up the crimp connection...would be a safeguard?
Its nice that you can solder and believe in it....but it has no place in fuel injection harnesses.
Do some research please into this please. You can buy a proper crimper...with dies to do every connection within your car just like factory....for between $60 and $100. Ray
- Joe Perez
- Posts: 44
- Joined: Thu Dec 04, 2008 2:26 pm
Re: Megasquirt + stock injection
Wow. I can't tell how much (if any) of that was aimed at me directly, but I've definately got my asbestos undies on right now.
The key thing, obviously, that the harness itself is rather rigidly mounted to the head in several places, so of course there is no flexation of the joints going on. That's the one killer for soldered wire-to-wire (or wire-to-board) connections.
(It's a joke, get it? The statement itself is a broad generalization.)
There are certainly a lot of applications where properly crimped connections can be superior to soldered connections. Low resistance isn't one of them, but they certainly can be more mechanically robust if done correctly.
Note how I keep referring back to "proper" and "correct." The sorts of crimp terminals that OEMs use are nothing at all like the insulated barrel connectors that most hobbyists have access to. The former have built-in strain-reliefs, fold-over crimps, etc. The latter have none of this.
Here's an example of the sort of crimp pin that the pros use:
Note how on the images at the very left, you can see that the tabs have been folded over and inwards (upwards in the image) and that there's a whole separate level of crimp going on just to grab the insulation for strain relief. These are examples of the sort of thing that distinguishes terminals of this kind from the more common barrel connectors, such as these:
I'm not saying that the latter are useless. Far from it, I use them every day both on my cars and in my professional life. I'm just saying that it's not equitable to compare them to the sort of wire terminations that OEMs use. These insulated terminals provide no strain-relief, have no positive wire retention (foldover) and simply cannot be assembled with the same level of precision.
I've been with PR&E (a division of the Harris Corp) since 1999. We build audio mixing consoles and also do full-service studio integration. To give you some sense of scale here, my most recent build out was CBS Radio's new HQ on Houston St in lower Manhattan. Spent about 2/3 of last year there building out 22 studios, two newsrooms, and a 60 rack TOC. So I know that of which I speak here.
This is an example of one of the more common crimp terminals that we use in the factory: http://mouser.com/ProductDetail/Molex/0 ... R0uxpus%3d We have two semi-automated workstations set up with these (one for male, one for female), where the operator just inserts the stripped end of the wire and hits a footswitch, and the machine locates, crimps, trims, and ejects the terminal in one operation. As you can see from the link, these terminals cost 1.6 cents each when purchased on reels of 75,000.
So, the terminals definitely aren't a cost driver. There's a high NRE in getting set up to do them, but that's amortized on a very long schedule, and at our company at least, is part of a general overhead budget that's not directly factored into the burdened cost of operating any one specific workcenter. And once all that's said and done, the labor required to make a crimped termination is trivial in the extreme. Like I said before, a semi-skilled wiring tech in our shop can do hundreds of these terminations per hour.
If we're talking about circuit boards, then yes. Modern SMC / reflow boards are almost 100% robot-assembled, and even the old through hole stuff was highly automated even before it got to the wave station.
But wire-end solder terminations? No such machine. Every shop I've ever been to still solders connectors onto cable with a hand-held iron.
Believe me. If it were cheaper to do soldered terminations, then you'd see a whole lot more of them out there, particularly in consumer electronics like stereo receivers. Those are actually a great example; take one apart, and you'll see that they typically solder the high-current lines which are sensitive to drop (eg: the connections to the MOSFET board, which is commonly a separate assembly), however everything else (the connection to the front and rear panels, etc) typically use crimp-style termination.
Open up your PC and look at the little cables that plug into the motherboard for things like the case fan, front panel switches / connectors, LEDs, etc. I guarantee you that every one of those is crimped. And this is bargain-basement stuff we're talking about here. If the Chinese factories that build those cable assemblies thought they could save one-eighth of a penny per unit by soldering those rather than crimping them, believe me- they'd be all over it.
I come across a lot of older electronics in my line of work. It's pretty common to see transmitters of 1970s vintage still humming away, and we run into a lot of test equipment (stuff like RF impedance bridges for tuning AM towers) that's much older than that- some, in fact, is of WWII vintage.
Now, you open this stuff up and you'll find all manner of terminations. In the real old stuff, pretty much everything was point-to-point soldered. In the 70s, you see a lot of wire-wrap. The transmitters in particular contain a lot of crimp connectors, many of which are the plastic-insulated barrel variety that we're familiar with, though of course they were all crimped with the expensive tools.
Up through the late 1950s, it was still pretty common to see cloth-jacketed cable inside these devices. One downside to this is that cloth is porous, so any of this stuff that was exposed to harsh environments is suffering pretty serious corrosion. It's not uncommon to open one of these machines up and find that the metallic portion of the cable (as well as the cloth insulation over it) have completely disintegrated and turned to dust, however the bit of solder that had been holding them to their terminal (as well as whatever portion of the wire was embedded in the solder) are typically 100% intact.
How's that for reliability?
First, I assume you're talking about °C rather than °F, as even conventional 50/50 solder doesn't flow until about 185°C (365°F), and the modern low-lead and lead-free stuff typically flows at around 220°C (425°F). These are minimum flow temps, your iron will typically run quite a bit hotter. The Weller WES51 which I use at home (a good, mid-range iron which you can buy for about $125) is adjustable between 177°C to 454°C.
And trust me, the folks designing solder terminals know this.
This idea that the temperatures encountered in a typical hand-soldering operation are going to cause any harm at all to the surface plating or locking tab or of an electronic terminal is just utter nonsense.
To give you one example, open up the fuel injector harness of any 1990 to 1993 Mazda Miata. These are four cylinder engines, port-injected, but with only two injector drive channels at the ECU. Within the harness, injector 1 & 3 are commoned together, as are 2 & 4, with the hot sides of all four commoned such that there are only three wires exiting the harness to the connector. These connections are all soldered in the factory injector harness. They are the only soldered harness connections I've yet encountered in that particular car, but it's an excellent example of an OEM soldered connection in an area that is subjected to vibration, heat, and so on.raygreenwood wrote:The endless debate. There has not been one single soldered connection in any production fuel injection system...ever.... for very good reasons. Vibration, heat, expansion, voids etc. etc. etc. etc. etc.
The key thing, obviously, that the harness itself is rather rigidly mounted to the head in several places, so of course there is no flexation of the joints going on. That's the one killer for soldered wire-to-wire (or wire-to-board) connections.
Broad generalizations are always wrong.No one cares how well you solder. Soldered connections are inferior in lifespan, resistance etc.
(It's a joke, get it? The statement itself is a broad generalization.)
There are certainly a lot of applications where properly crimped connections can be superior to soldered connections. Low resistance isn't one of them, but they certainly can be more mechanically robust if done correctly.
Note how I keep referring back to "proper" and "correct." The sorts of crimp terminals that OEMs use are nothing at all like the insulated barrel connectors that most hobbyists have access to. The former have built-in strain-reliefs, fold-over crimps, etc. The latter have none of this.
Here's an example of the sort of crimp pin that the pros use:
Note how on the images at the very left, you can see that the tabs have been folded over and inwards (upwards in the image) and that there's a whole separate level of crimp going on just to grab the insulation for strain relief. These are examples of the sort of thing that distinguishes terminals of this kind from the more common barrel connectors, such as these:
I'm not saying that the latter are useless. Far from it, I use them every day both on my cars and in my professional life. I'm just saying that it's not equitable to compare them to the sort of wire terminations that OEMs use. These insulated terminals provide no strain-relief, have no positive wire retention (foldover) and simply cannot be assembled with the same level of precision.
I can only assume that you've never worked in an electronics manufacturing environment.Actually...crimping is about four times MORE expensive than soldering....because of teh cost of the connector alone. The terminals alone are extremely expensive. Factories quit soldering ANY terminals other than hard connections to boards...decades ago..because of reliability issues.
I've been with PR&E (a division of the Harris Corp) since 1999. We build audio mixing consoles and also do full-service studio integration. To give you some sense of scale here, my most recent build out was CBS Radio's new HQ on Houston St in lower Manhattan. Spent about 2/3 of last year there building out 22 studios, two newsrooms, and a 60 rack TOC. So I know that of which I speak here.
This is an example of one of the more common crimp terminals that we use in the factory: http://mouser.com/ProductDetail/Molex/0 ... R0uxpus%3d We have two semi-automated workstations set up with these (one for male, one for female), where the operator just inserts the stripped end of the wire and hits a footswitch, and the machine locates, crimps, trims, and ejects the terminal in one operation. As you can see from the link, these terminals cost 1.6 cents each when purchased on reels of 75,000.
So, the terminals definitely aren't a cost driver. There's a high NRE in getting set up to do them, but that's amortized on a very long schedule, and at our company at least, is part of a general overhead budget that's not directly factored into the burdened cost of operating any one specific workcenter. And once all that's said and done, the labor required to make a crimped termination is trivial in the extreme. Like I said before, a semi-skilled wiring tech in our shop can do hundreds of these terminations per hour.
I guess I've never seen one of these soldering machines. Funny, you'd think I'd have come across at least one, given the number of assembly houses I've been in.All soldered connections from the late 60's on...were machine soldered.
If we're talking about circuit boards, then yes. Modern SMC / reflow boards are almost 100% robot-assembled, and even the old through hole stuff was highly automated even before it got to the wave station.
But wire-end solder terminations? No such machine. Every shop I've ever been to still solders connectors onto cable with a hand-held iron.
I never said it was started as a money saver. But that is definitely one of the major drivers in its use today. The reality is that an operator can turn out five to ten times the number of crimped connections per hour vs. soldered connections, and apart from the initial cost of the tooling, the parts cost of crimp pins, as I noted previously, is insignificant.This myth that crimping was started as a money saver...is BS.
Believe me. If it were cheaper to do soldered terminations, then you'd see a whole lot more of them out there, particularly in consumer electronics like stereo receivers. Those are actually a great example; take one apart, and you'll see that they typically solder the high-current lines which are sensitive to drop (eg: the connections to the MOSFET board, which is commonly a separate assembly), however everything else (the connection to the front and rear panels, etc) typically use crimp-style termination.
Open up your PC and look at the little cables that plug into the motherboard for things like the case fan, front panel switches / connectors, LEDs, etc. I guarantee you that every one of those is crimped. And this is bargain-basement stuff we're talking about here. If the Chinese factories that build those cable assemblies thought they could save one-eighth of a penny per unit by soldering those rather than crimping them, believe me- they'd be all over it.
For a properly done machine-crimp, they're pretty good. Hermetically-sealed is a misnomer unless you apply a conformal coating, but they're good. For the round, insulated barrel connectors that most folks have access to, not even close.Did you know that properly crimped terminals are hermetically sealed? The gap between terminal and wire is non-existant.
Yes, it's the nature of the bond between solder and wire. If yours aren't, then you're doing it wrong.Can you guarantee that with with soldering?
I come across a lot of older electronics in my line of work. It's pretty common to see transmitters of 1970s vintage still humming away, and we run into a lot of test equipment (stuff like RF impedance bridges for tuning AM towers) that's much older than that- some, in fact, is of WWII vintage.
Now, you open this stuff up and you'll find all manner of terminations. In the real old stuff, pretty much everything was point-to-point soldered. In the 70s, you see a lot of wire-wrap. The transmitters in particular contain a lot of crimp connectors, many of which are the plastic-insulated barrel variety that we're familiar with, though of course they were all crimped with the expensive tools.
Up through the late 1950s, it was still pretty common to see cloth-jacketed cable inside these devices. One downside to this is that cloth is porous, so any of this stuff that was exposed to harsh environments is suffering pretty serious corrosion. It's not uncommon to open one of these machines up and find that the metallic portion of the cable (as well as the cloth insulation over it) have completely disintegrated and turned to dust, however the bit of solder that had been holding them to their terminal (as well as whatever portion of the wire was embedded in the solder) are typically 100% intact.
How's that for reliability?
No, it doesn't. In fact, terminals intended for solder-termination typically use the same materials and plating techniques as their crimp-intended counterparts.Also...soldering creates brittleness in the very thin connector systems of the EFI harnesses. Try heating one to anything over about 350 and see what happens. Been there and done that. The plating systems and level of spring in the terminals of EFI harnesses are critical. Heat kills that.
First, I assume you're talking about °C rather than °F, as even conventional 50/50 solder doesn't flow until about 185°C (365°F), and the modern low-lead and lead-free stuff typically flows at around 220°C (425°F). These are minimum flow temps, your iron will typically run quite a bit hotter. The Weller WES51 which I use at home (a good, mid-range iron which you can buy for about $125) is adjustable between 177°C to 454°C.
And trust me, the folks designing solder terminals know this.
This idea that the temperatures encountered in a typical hand-soldering operation are going to cause any harm at all to the surface plating or locking tab or of an electronic terminal is just utter nonsense.
Link?You can buy a proper crimper...with dies to do every connection within your car just like factory....for between $60 and $100. Ray
- raygreenwood
- Posts: 11895
- Joined: Wed Jan 22, 2003 12:01 am
Re: Megasquirt + stock injection
Joe Perez wrote:Wow. I can't tell how much (if any) of that was aimed at me directly, but I've definately got my asbestos undies on right now.
Actually none of that was aimed at anyone in particular or was meant to be a sleight or cutting.
To give you one example, open up the fuel injector harness of any 1990 to 1993 Mazda Miata. These are four cylinder engines, port-injected, but with only two injector drive channels at the ECU. Within the harness, injector 1 & 3 are commoned together, as are 2 & 4, with the hot sides of all four commoned such that there are only three wires exiting the harness to the connector. These connections are all soldered in the factory injector harness. They are the only soldered harness connections I've yet encountered in that particular car, but it's an excellent example of an OEM soldered connection in an area that is subjected to vibration, heat, and so on.raygreenwood wrote:The endless debate. There has not been one single soldered connection in any production fuel injection system...ever.... for very good reasons. Vibration, heat, expansion, voids etc. etc. etc. etc. etc.
The key thing, obviously, that the harness itself is rather rigidly mounted to the head in several places, so of course there is no flexation of the joints going on. That's the one killer for soldered wire-to-wire (or wire-to-board) connections.
Broad generalizations are always wrong.No one cares how well you solder. Soldered connections are inferior in lifespan, resistance etc.
(It's a joke, get it? The statement itself is a broad generalization.)
....like a person with good humor!
There are certainly a lot of applications where properly crimped connections can be superior to soldered connections. Low resistance isn't one of them, but they certainly can be more mechanically robust if done correctly.
Actually lower LONG TERM reistance and potential IS one of the benefits. The fact that there is no way to guarantee void free solder at the high rate of production of crimped connectors, or to guarantee corrosion free fluxless solder because of the make up of low temp solder...crimped connectors with dedicated plating materials are superior in every way to solder...longterm...in high vibration, variable heat and high humidty COMBINED environments
Note how I keep referring back to "proper" and "correct." The sorts of crimp terminals that OEMs use are nothing at all like the insulated barrel connectors that most hobbyists have access to. The former have built-in strain-reliefs, fold-over crimps, etc. The latter have none of this.
Here's an example of the sort of crimp pin that the pros use:
I wouldn't consider Molex terminals proffesional for Automotile. Molex round pins have too many failure modes in both Fuel injection and in air rcaft in resistance sensitive environments....which is why dual cantilever's are used in most of these. There is plenty of history and documentation to support that. Just ask GM. But I agree on the crimps. All of my crimping tools and dies....as stated earlier...are highly proffesional (Palladin)
Note how on the images at the very left, you can see that the tabs have been folded over and inwards (upwards in the image) and that there's a whole separate level of crimp going on just to grab the insulation for strain relief. These are examples of the sort of thing that distinguishes terminals of this kind from the more common barrel connectors, such as these:
We agree whole heatedly on crimp quality. Nice Diagrams
Then I'll say it. Those types of terminals are poor for automotive use other than basic chassis terminations. I use them nowhere at all except for stationary industrial applications and machines in closed cabinets. I use all factory compound crimp terminals with integral strain relief....just like in your diagram
I'm not saying that the latter are useless. Far from it, I use them every day both on my cars and in my professional life. I'm just saying that it's not equitable to compare them to the sort of wire terminations that OEMs use. These insulated terminals provide no strain-relief, have no positive wire retention (foldover) and simply cannot be assembled with the same level of precision.
ou
I can only assume that you've never worked in an electronics manufacturing environment.Actually...crimping is about four times MORE expensive than soldering....because of teh cost of the connector alone. The terminals alone are extremely expensive. Factories quit soldering ANY terminals other than hard connections to boards...decades ago..because of reliability issues.
I've been with PR&E (a division of the Harris Corp) since 1999. We build audio mixing consoles and also do full-service studio integration. To give you some sense of scale here, my most recent build out was CBS Radio's new HQ on Houston St in lower Manhattan. Spent about 2/3 of last year there building out 22 studios, two newsrooms, and a 60 rack TOC. So I know that of which I speak here.
I am always impressed with people who work in the lectronics industry...but none or few of those connector systems you illustrated must live up to the vibration, weathering and heat cycling of automotive or aircraft terminals where dual cantilever terminals are spec'd. Different applications altogether.
This is an example of one of the more common crimp terminals that we use in the factory: http://mouser.com/ProductDetail/Molex/0 ... R0uxpus%3d We have two semi-automated workstations set up with these (one for male, one for female), where the operator just inserts the stripped end of the wire and hits a footswitch, and the machine locates, crimps, trims, and ejects the terminal in one operation. As you can see from the link, these terminals cost 1.6 cents each when purchased on reels of 75,000.
So, the terminals definitely aren't a cost driver. There's a high NRE in getting set up to do them, but that's amortized on a very long schedule, and at our company at least, is part of a general overhead budget that's not directly factored into the burdened cost of operating any one specific workcenter. And once all that's said and done, the labor required to make a crimped termination is trivial in the extreme. Like I said before, a semi-skilled wiring tech in our shop can do hundreds of these terminations per hour.
Those terminals in your diagram......the springless, unplated round molex terminals are dirt cheap....and...they will never even come close to performing as well as the dual cantilever terminal system in this environment. Those were tested by AMP and Bosch right along side D-jet terminals in 1978. They failed in marginally the same number of cycles as the D-jet system....because they have no secondary spring.
Once the natural spring tension of the barrel fatigues through plugging and unplugging cycles.....or heat cycles or vibrational wear...they are junk for severe duty connectivity. The only place these do do well....are when they are encased in a guided and sealed screw type, male/female, mil-spec housing....far too expensive and bulky for automotive. The dual cantilever, brass, bronze or phosphor bronze terminal, with nickle plated cantilver spring...with or without selective gold plate at either 8 or 10 micron...are about 10-12 times that cost.....in roll quantity.
You will find that I am well researched in my terminals. Cost is very high on proper engineered terminals....but has nothing to do with it when performance....and ...system warrantability far exceed un-spring terminals by several hundred percent. Terminal cost is not a cost driver....warrantability and stability are.....since that was found to be the original and main problem in automotive fuel injection terminals from the inception of D-jet in 1966 through the ramping up of airflow based L-jet in 1973-74.
I guess I've never seen one of these soldering machines. Funny, you'd think I'd have come across at least one, given the number of assembly houses I've been in.All soldered connections from the late 60's on...were machine soldered.
You need to get out more. I work in the printed electronics industry. Having been in over 3200 plants....a little over 300 of them in a wide range of electronics ranging assembly from solar to automotive...I have seen plenty of them. Most were obsolete by the late 50's to mid 60's and sitting around collecting dust and being used for repairs on old systems. Almost all were custom built jig assemblies. The latest I ever saw were some of the earliest pick and place machines with solder capability that still needed loading all of the wires into ferrules by the operator.
If we're talking about circuit boards, then yes. Modern SMC / reflow boards are almost 100% robot-assembled, and even the old through hole stuff was highly automated even before it got to the wave station.
But wire-end solder terminations? No such machine. Every shop I've ever been to still solders connectors onto cable with a hand-held iron.
Again....you spent too much time in rigid board. You need to get around more. I'm talking automotive secondary and sub-harnesses here
I never said it was started as a money saver. But that is definitely one of the major drivers in its use today. The reality is that an operator can turn out five to ten times the number of crimped connections per hour vs. soldered connections, and apart from the initial cost of the tooling, the parts cost of crimp pins, as I noted previously, is insignificant.This myth that crimping was started as a money saver...is BS.
Believe me. If it were cheaper to do soldered terminations, then you'd see a whole lot more of them out there, particularly in consumer electronics like stereo receivers. Those are actually a great example; take one apart, and you'll see that they typically solder the high-current lines which are sensitive to drop (eg: the connections to the MOSFET board, which is commonly a separate assembly), however everything else (the connection to the front and rear panels, etc) typically use crimp-style termination.
It has always been cheaper to do soldered connections primarily because the slower output still well filled the demand...and as I noted, many of the sub-sassembly harness making systems were mass solder jigs. It was still so cheap to solder when EFI hit the market....that if the performance had been with warrantable tomerance...they would still be doing it. As it turned out....even the first terminal system they used......was so bad that soldering probably would have been an improvement. bear in mind that solder in kind when going from a flexible to a rigid surface.....(wire to PC board)...but cannot handle the long term, vibration, expansion etc...of a moving wire loom.
Open up your PC and look at the little cables that plug into the motherboard for things like the case fan, front panel switches / connectors, LEDs, etc. I guarantee you that every one of those is crimped. And this is bargain-basement stuff we're talking about here. If the Chinese factories that build those cable assemblies thought they could save one-eighth of a penny per unit by soldering those rather than crimping them, believe me- they'd be all over it.
As I noted above....what you connect to a PC board has zilch to do with what happens in a flexible harness ina 200+ degree environment, with solvents, oils, water and vibration in the mix. And....EFI boards still have soldered connections and wires within the enclosure.....this is not what any of this thread had to do with.
For a properly done machine-crimp, they're pretty good. Hermetically-sealed is a misnomer unless you apply a conformal coating, but they're good. For the round, insulated barrel connectors that most folks have access to, not even close.Did you know that properly crimped terminals are hermetically sealed? The gap between terminal and wire is non-existant.
Yes, it's the nature of the bond between solder and wire. If yours aren't, then you're doing it wrong.Can you guarantee that with with soldering?
And unfortnately...you cannot guarantee every solder joint at thousands of terminations per hour (machine or hand done), nor can you guarantee them to be voidless. All it takes is one bad solder joint...that may not even go out until long after the car leaves the factory and the harness leaves the bench....but it is now a warranty issue. But you can do this with crimping
I come across a lot of older electronics in my line of work. It's pretty common to see transmitters of 1970s vintage still humming away, and we run into a lot of test equipment (stuff like RF impedance bridges for tuning AM towers) that's much older than that- some, in fact, is of WWII vintage.
All of which is true..and has has nothing to do with flexible, vibrating, high temp harnesses in oily, dustry, humid environment. I have a 1930 something telephone and RCA radio full of solder and still operating.....none of which have any of the constraints or demands of an automotive fuel injectionsystem....which is all we are talking about here.
Now, you open this stuff up and you'll find all manner of terminations. In the real old stuff, pretty much everything was point-to-point soldered. In the 70s, you see a lot of wire-wrap. The transmitters in particular contain a lot of crimp connectors, many of which are the plastic-insulated barrel variety that we're familiar with, though of course they were all crimped with the expensive tools.
Up through the late 1950s, it was still pretty common to see cloth-jacketed cable inside these devices. One downside to this is that cloth is porous, so any of this stuff that was exposed to harsh environments is suffering pretty serious corrosion. It's not uncommon to open one of these machines up and find that the metallic portion of the cable (as well as the cloth insulation over it) have completely disintegrated and turned to dust, however the bit of solder that had been holding them to their terminal (as well as whatever portion of the wire was embedded in the solder) are typically 100% intact.
How's that for reliability?
Perfect...beautiful....Viva Solder! I have always been a fan....but what if anything does that have to do with the proven UNRELIABILITY of soldered connections in electronic fuel injection systems. Its kind of like you are worrying about closing the stable doors after the horse has crapped and left. Vast amounts of testing and research have already proveen beyond the shadow of a doubt that soldered connections do not hold up..... to a warrantable stable resistance level in the hostile environment of automotive fuel injection. This is why not ONE single company of mass production vehicles world wide has EVER used it in fuel injection harnesses. Just name one. I'll get to teh Mazda below.
No, it doesn't. In fact, terminals intended for solder-termination typically use the same materials and plating techniques as their crimp-intended counterparts.Also...soldering creates brittleness in the very thin connector systems of the EFI harnesses. Try heating one to anything over about 350 and see what happens. Been there and done that. The plating systems and level of spring in the terminals of EFI harnesses are critical. Heat kills that.
First, I assume you're talking about °C rather than °F, as even conventional 50/50 solder doesn't flow until about 185°C (365°F), and the modern low-lead and lead-free stuff typically flows at around 220°C (425°F). These are minimum flow temps, your iron will typically run quite a bit hotter. The Weller WES51 which I use at home (a good, mid-range iron which you can buy for about $125) is adjustable between 177°C to 454°C.
Nope: and you hit it on the head in your statement. In general.....MOST solder does not flow until it is a hair under 400F (not quite true I have 225-250F solder laying on my desk right now).....but that solder will not have anywhere near the strength required to use a solder terminal in anything because of the alloy....anywhere but the most protected environment.....like attached to a board and then encapsulated.
My point in the above statement...you made even better for me. The point was that the the required small terminals for EFI with thin metal gauge and delicate engineered platings.....will not survive much over 300F even with heat sinks. Been there and done that. They distort too much to make contact that is secure. So you might say ....then go to a dedicated solder terminal that has no heat problems? They do not make any in the gauge and type that work with the unified range of fuel injection components used world wide...and certainly no dual cantilever system's spring would survive the heat......that is the point I was making. Soldering dual cantilevered terminals and even rolled spring female terminals....will wreck the spring temper of either rendering them worthless in the short term
And trust me, the folks designing solder terminals know this.
This idea that the temperatures encountered in a typical hand-soldering operation are going to cause any harm at all to the surface plating or locking tab or of an electronic terminal is just utter nonsense.
Not true at all. They distort the terminals of modern timer style and dual cantilever terminals....badly. Because of expansion and the thin micron platings...the platings crack. If you get bored I'll send you some microscope shots. These cracks promote corrosion inside the cracks and wear through via abrasion on the opposite terminal.
Link?You can buy a proper crimper...with dies to do every connection within your car just like factory....for between $60 and $100. Ray
It appears you have alot of experience. Impressive! But I think you got a little hung up on the solder versus crimp issue....but missed the one qualifying point of this whole thread. This is about severe location automotive (and many aircraft), high and low temp, wet, dry, vibration and chemical prone environment wiring harnesses. What happens in automotive chassis harness (until the last decade) is almost apples to oranges as well because there were few logic-fed, interogative, resistance based wires in the chassis harness. In many cases you could solder to your heats content Get it?! .
There were few times when the changes in connectivity, resistance or potential...would do more than screw up your dash lights or stop your stereo or air conditioner. Not so any more. With virtually everything PC controlled in the chassis harness....virtually all solder has been replaced by much more reliable crimped AND cantilevered, vibration protected connectors in chassis harness as well.
Also a note on the Miata, being that comparatively few were built (compared to other models) I would not be surprised to find soldering as you did...and....the Miatas injection is also not the model of reliability for high milage anyway. I have worked on numerous Miata...and not all of them were done that way. At least I have not found that solder joint....but you learn something every day.
Also, one of the largest users of AMP designed, patented and licensed built dual cantilevered terminals....are ALL of the Japanese car makers. They were early adopters of the system.
Also...just so you know I was not poking at you or anyone else at all. Its just if for no other subject....when it comes to fuel injection terminals...I know exactly what I am doing. I have been working with and researching...fuel injection terminals and OEM systems.....since about 1979. Very nice yakking with you! Ray
-
- Posts: 3944
- Joined: Sun Apr 27, 2008 3:36 pm
Re: Megasquirt + stock injection
raygreenwood wrote:{SNIP}
Good Lord, snip the post. Pet peeve, I have to scroll through a long ass post that someone else just posted.
-
- Posts: 3391
- Joined: Tue Sep 03, 2002 12:01 am
Re: Megasquirt + stock injection
So if we break this down...
Ray has never seen a factory FI soldered joint...but Joe has and pointed out where...Ray has worked on numerous of said cars and did not know of such a joint, insisting all joints were crimped. So it's safe to say Ray has never seen a factory FI solder joint fail, ever, and that any failures he has seen (if any) have been of the crimped variety.
Ray has never seen a factory FI soldered joint...but Joe has and pointed out where...Ray has worked on numerous of said cars and did not know of such a joint, insisting all joints were crimped. So it's safe to say Ray has never seen a factory FI solder joint fail, ever, and that any failures he has seen (if any) have been of the crimped variety.
-
- Posts: 23
- Joined: Tue Jan 12, 2010 7:17 pm
Re: Megasquirt + stock injection
I think the topic of this should be changed ... Although some good discussion here.
'78 Super Convertible restoration in progress!