bobclevenger

I'd be willing to bet that this is what the PIE unit does. If anyone wants to try this directly, be aware that you do NOT want your audio source grounded to the car -- even through a power supply. This is because the c.d.e.&f pins are balanced inputs and not grounded. A permanent solution using these pins (if they do work) will have to include some sort of DC isolation - (transformer or capacitor coupling).

Let's not have anyone breaking their stereo here.

I am tending to believe that somehow PIE made a run of adapters that accidentally provide the "trigger" to enable the aux-in. Most likely one of the "unused" pins is going to ground.

Thew, you seem to have established a line of communication to Logjam and PIE. Can you ask them this:
Would PIE like to examine a unit that works with RX-8s? I would be willing to send them mine to examine.

Also, to narrow this down, if anyone who has a PIE adapter set that does not work is willing to send it to me, I will put it in my 8 and see if it works or not. That would nail down whether it's the PIE or the RX-8 that's causing the problem. If it works on mine and not on yours, then we know that is is a difference in rx-8s; if it doesn't work on yours or mine, then we know that it's a difference in PIE adapters. I could also see if it's the HU to Mbus adapter or the Mbus to RCA adapter that is the culprit (as long as I still have mine with which to compare, that is).

truemagellen

hey guys with Nav (that includes me ) I found this in the 'attention ipod owners' thread:

So is there a separate wiring diagram for nav owners??? and may explain the Factory only option?

Thew the wiring diagram posted by MazdaManiac may be for the non-nav owners only but I'm not really sure

Sputnik

Folks,

Let's keep this thread on topic, meaning keep posts regarding the PIE adapter, good, bad, or indifferent. This thread is long enough that we don't need to waste space telling people how another solution works better.

Thank you.

---jps

thew

ok bob .. lets talk... send me yours.. or ill send you mine... I Must figure this out.. in the mean time i may send some signel down the line to the HU and see what happens.

bobclevenger

Answered in PM.
It occurs to me that the most likely answer to why some PIE units work and others do not is component tolerances. Some of them can be pretty large.

I'm hoping that it's on the PIE side of things, because then PIE can just change a component value (and it would be to their advantage to do so since we are a pretty good-sized market).

eXentric

I'm trying to keep the technical "hacking" of the PIE module separate from this thread discussing the module. Since there is already another thread based on hacking the rear port, I've posted my thoughts there.

I have an idea for the people who have the unit cutting out on them. If your unit isn't working as expected, please see this thread:

Engineers: Can this be hacked

eXentric

I believe the original problem may be in the protocol and that it would be possible for PIE to change it. After looking inside the module, I can see that there is a programmable PIC chip in there which is where the protocol conversion is done. That chip can most likely be repogrammed, but you have to have the right tools. Even if PIE fixed the problem, you would most deffinately have to send the unit back to them to get that chip swapped out or reprogrammed.

I came up with a new theory today (discussed in the other thread I linked to). Now I'm wondering if it is the head unit or the PIE module that is causing the sound to be muted...

eXentric

I guess I could have just posted those thoughts in this thread. I was thinking the other thread to be more of a technical / hacking thread then this one. Regardless, we can discuss it here or there. I'd just like to see this thing work. I'll have to test mine Sunday because I'm much too busy the next two days.

thew

hehe.. ive got them all so it makes no diff..

I am working with Bob will let you know what we find.

martinl78

Called PIE tech support. They insist that the adapter will work, but with the RCA option that it isn't expected to work. They suggested that they are working on a one piece component that should be available in a couple of months. [don't know how much faith to put in that.]

he suggested I return the cables I purchased to the places I got them from.

???


Martin

thew

try spliting the inputs..

Tamas

iTrader: (1)

Same here - I talked to a guy called Victor and he confirmed what was said here earlier: they only tested the RCA cable / Alpine adapter in the 2002-2004 MPV minivan. He told me that they are working on a solution but it definitely will be a separate device and not an updated PC-ALP module. They won't provide a fix or a hack to make these curent modules work. So if somebody's unit works, great - but there is no guarantee. Apparently they also told Logjam to change the description on their site so that it won't say that this combo will work with all the models the PC-ALP otherwise is compatible with.

RX8J

I have no NAV. Not sure of the build date of the car I would have to look and it's late right now. I will post.

I think if we are going to get to the bottom of this, we will have to find two people in one area one that works, and the other that does not. That way those people could switch out components and verify if it's a PIE problem, or a radio problem. Until someone does that, it's going to be all speculation.

-RX8J

starbucks

It's beyond my understanding that the PIE set up would work on some with the configuration -- single cd, cd changer, nav, no nav, bose, no bose, etc. -- and not on others of the same configuration. That doesn't make any sense. Either it should work or not. I can understand an occasional bad part, but not on the scale that appears to be the case if you read all the posts -- works for some, doesn't for others with the same set ups.

I just don't get it.

martinl78

I just went to try using a different device as the player that had both headphone out and line out. Line out was the way to go. However...still didn't work.

I also tried with the screen open this time instead of closed. Sure enough, when EX 1 stops working, something kicks in and takes over the display from the NAV unit and puts a blue screen up that says TV 1 and a bunch of Japanese on one line. I found that if you press the MENU button on the NAV control, it gives you a submenu to ADJust the screen brightness and contrast. It will say Day Mode. If you turn on the headlights, it then says night mode and will allow you to customize brightness and contrast for nighttime separately.

It will also allow you to turn the screen off. It does seem to turn on again later when you kick it into the blue screen. SO -- now we know of a way to tweak the display settings at least.

Also, if while in this mode you press POS, it will exit this mode and go to NAV.

However, this seems to have no effect on the EX 1 issue.

Also found that if you press the Disc up or down button to change the display to EX 2 through 6, suddenly the sound goes off. I wonder if it is accepting input through one of the unused pins on the connector?

I also noticed that using the MUTE button does mute the sound as expected but doesn't prevent the TV 1 screen or anything.

Anybody think Mazda might tell us anything?

Also, the TV 1 screen seems to have 6 buttons across the bottom, and one that says AUTO-M. I can't help but wonder if there is a infra-red remote since the LCD screen looks like it might have a receiver window for one. Makes me wonder what other settings one might be able to access if there were such a thing.

martinl78

Also...makes me wonder...for those that have this working, do you get the TV 1 screen with Japanese? (Assuming this really does work with a NAV package installed...I forgot if anyone had that combo working with Bose.)

Thanks.
Martin

bobclevenger

Mine is Nav and Bose and my PIE input is working.
When it quit working for a while the other day I had the TV 1 screen with Japanese characters on the display, but not when it is working. It just shows my nav screen or my DVD/TV unit or my reverse camera.

bobclevenger

Thew and I are doing exactly that. Stay tuned for updates.

bobclevenger

BTW, I tried my PIE aux-in with ALL the video sources completely disconnected today and it made no difference at all.

bobclevenger

Update time.
The PIE adapter set works in my RX-8 as you all know if you've been following this thread. I just finished pulling my PIE parts and replacing them with Thew's set which doesn' work in his 8. They work just fine in mine. They have been playing for over 5 minutes now.

So, what's the difference in our 8s? Well, mine has the mp3 player and nav, so the presence of nav isn't the problem as was suspected at one time. I just wonder if the mp3 player is sending a signal that enables the aux-in in the HU? Perhaps a signal that remote CD changers send? The factory CD changer doesn't use this connector, so it might not provide that signal. But the connector is for the JDM TV module which WILL work with auxiliary devices such as the RCA cable that Thew got from Japan -- so a signal saying "I'm a remote device and I'm here" is probably needed at the port we are using. The Alpine changers apparently provide this signal. I'm wondering if my mp3 player is also providing the same signal even though it isn't using the 16-pin connector (it's probably on the same data bus).

Enough speculation. I've worked my curiosity up enough to pull the mp3 player off of the HU and see if the PIE stuff still works.

Further news as it happens.

bobclevenger

I'm really confused now.
I pulled the mp3 module from the HU and installed the bare HU in the car. Powered it up, and the PIE aux-in was still working like a charm! I let it play for 5 minutes and all was well. So it's not the mp3 player that's allowing the PIE stuff to work. It must be some difference in the HUs. As I said before, it's not the presence or absence of a video signal, because I tried it both ways. And it worked both ways.

At this point I'm going to put my HU back together, put it back in my car, reprogram my radio stations, and pack Thew's PIE parts back up to ship back to him tomorrow.

Thanks for the loan of the parts Thew. We at least narrowed things down a bit, but we still don't have a fix for you and the rest whose PIE adapters don't work right.

Anybody else see anything that I'm missing here?

thew

i assume you tried my PIE with and without your RCA spliters ?

and this was with and without my wireing harness?

just trying to get my head around it.. I really have a had time beliveing that your HU is not the same as mine..

What build day is your 8 ?
Mine is 04/03

bobclevenger

My build date is 06/03. Not much difference.

Yeah, this was with and without multiple inputs; with your harness; with my harness; I tried every way I could think of to make it fail, and it wouldn't. I even marked my PIEces so that I wouldn't get yours confused with mine.

I'm beginning to believe in magic!

Oh well, at least I accomplished one thing. I got my PIE box mounted to the underside of the HU (in the space for the cassette tape or MD module) with heavy-duty mounting tape. Better than where it was.

I'm beginning to wonder if some 8s were made with MPV HUs? I don't even know if they'd be physically compatible or not -- probably not.

truemagellen

wait mine is 04/03 there were some changes in efficiency for creation of parts often during 2003 out at the Mazda Plant in Hiroshima (but is the HU built in Hiroshima?)

this just got really really confusing...thanks Bob for spending the time to figure this out

eXentric

Got mine installed yesterday and it does work. I too assume that it is a head unit issue. Here's my logic:

On mine, it does say "EX 1" but the "1" flashes all the time. This says to me that the head unit is waiting for the CD changer to load up disc 1 (which it obviously cannot do). When I first talked to a technical contact at PIE, he told me that the module could not be used for AUX purposes because the logic chip in the PIE unit is only a translator. He said that without the CD changer connected, there is nothing to translate. What this means is that the head unit sends the Mazda command to select disc 1. The logic chip in the PIE module translates this into an Alpine command and sends it out the port. But there is no hardware (CD Changer) to receive the command and no hardware to send the Alpine "Change Successful" response back. If there was, the Alpine "Change Successful" command would be translated back to the Mazda "Change Successful" command and the "1" in "EX 1" would stop flashing.

Apparently, some head units in some of our cars are 'lazy' and just let this slide. Some other units, however, see that the CD has not changed after an amount of time and decide to cut the audio. This is a somewhat logical move to make because if the changer is malfunctioning, the audio signal it’s sending is likely to be invalid.

There are two ways to address this, but unfortunately neither is something we will be able to do ourselves. The first place it can be resolved is in the head unit logic. I have my doubts that the head unit logic can even be physically updated (the chip is probably not flashable), so the only option is to swap out the head unit. The second option is to update the microchip in the PIE unit, which does in fact seem to be flashable. The update would have to send a Mazda "Change Successful" command BEFORE the timeout occurred, even if no Alpine changer was found. The code on this chip then becomes more then merely a simple "translator".

This COULD be done, but for someone to try and start from scratch would be a very daunting task. With the right tools, one could probably extract the existing program off the chip in assembly. With that in hand and enough understanding of the protocols, the program could be modified and the chip re-flashed. But the only person I know that remotely has enough knowledge to do this is OverLOAD, and I doubt he would take the time to do it. Not because he wouldn't want to help but because not only would he not profit from it, he would also be helping a competing product.

If someone else wanted to attempt this, I think it would be somewhat straight forward (for someone that understands assembly and microcode). Since the program on the chip is a simple translator, I would expect there to be two lookup tables. The first table would hold a list of commands that come in from the head unit and the respective Alpine command to send back out. The second table would be the list of responses coming from the Alpine changer and the associated Mazda response to send back to the head unit. Any commands and responses that are not needed would be ignored and therefore this list would be fairly short. In fact, I would venture to guess that there are probably only 3 or 4 entries in the response table (one of them being the “Change Successful” command I mentioned about earlier).

As you can tell, I’ve dabbled in this stuff but I am by no means an expert. It is my opinion that this is completely within the realm of possibility for PIE to fix. I would also guess that the have already fixed it and are getting ready to release a version dedicated to AUX input. This would probably be a different beast however: it wouldn’t be a translator like the current module but instead simply a protocol “faker” that looks only for “Change to Disc 1” and responds with “Change successful”. I think it would be extremely simple for them to do and I would guess it’s on its way.

We could compare serial numbers, model numbers, manufacture plant information, etc. of our head units to prove it’s the head unit. But I suspected it was from the moment I plugged mine in --- even before I read that thews module worked for bob.

This is unfortunate guys and very frustrating to have some of the knowledge but not enough to help. I’m very sorry for those of you that it doesn’t work for. It’s even more unfortunate that you may be past your return date. If you’re not, I strongly suggest you return it and see what comes from PIE. Our other option is to contact PIE and ask them if they are working on an AUX module and if it would be possible to get our current module reprogrammed with that code (assuming such a thing exists). But I don’t have high hopes for them cooperating in that way.

Best of luck to all,

eXe