[86GT]

What is weird about that is it will start to log and then goes nuts.

If you slow the polling down to say 500 ms what does it do.

[Dale McPeters]

What is weird about that is it will start to log and then goes nuts.

If you slow the polling down to say 500 ms what does it do.

I will double check that this afternoon as I also have some stuff to check for Paul.

[Craig Moates]

I just talked at some length with Dale, and it sounds to me like he is suffering from the USB cable shortcomings similar to other folks. That will be addressed in very short order, the parts will be here tomorrow and I'll begin shipping out to get everyone updated.

[cgrey8]

In the meantime, is there a way to force USB1.1 so it doesn't attempt a USB2.0 connection?

[Craig Moates]

I don't think so, it's more of a matter that the USB 2.0 spec is more stringent in terms of cable quality.

My vendor snuck some 1.0/1.1 cables in on me and that seems to be the root cause at this point.

[Dale McPeters]

Clint I have the QH up and running in the '93 cobra with Paul's software....data logging and all.

I will drop you an email off line.....

Man is this thing fast!!!!

Craig thanks for such excellent hardware support!!!!

[area302]

Hello all,
I had the same problem, I installed my digital camera cable with a ferrite bead and it worked great!!
I then installed a ferrite bead on the cable Craig supplied me, and it died the same way.
I thought it was a bad cable, but in my normal obsession to fix things, I picked up a large ferrite bead from Radio Shack, large enough to DOUBLE LOOP the cable through it.
It works perfect, runs all day with np problems now!!!
This is a large ferrite bead, so it is installed just outside the EEC.
Remember to DOUBLE LOOP it through, a single will not work.
Also FYI to everyone, check this link out:

http://www.k0bg.com/beads.html

Now we need a support group so I will stop tuning my engine constantly...lol

[cgrey8]

Did you note the impedance rating of the ferrite bead you used? I ask because there's an article I read that talks about the differences between USB1.1 and USB2.0 as it relates to the size of ferrite bead you put on the wire:
End USB Noise Without Stopping the Bus

The gist of the article is that standard ferrite beads are ineffective on USB2.0 signals. Instead what's needed is a common-mode choke coil. But even that has to be the right impedance to work with USB2.0 frequencies. Too much impedance from the choke can actually hurt USB2.0 where the same choke might help USB1.1. In USB1.1 communications, the frequencies of legitimate communication are well below the noise levels that can get introduced onto the wire. And thus putting a higher impedance choke on the wire is more effective at cleaning the noise out of the signal. But the frequencies of USB2.0 communication happen to be very close to the same frequencies as noise. So if you filter too much, you wind up filtering legitimate signals.

I did confirm that the non-USB2.0 extension cable I had on my QH was the main problem with frequent errors and drop-offs of the QH for me...I just added it about 2 weeks ago and it didn't occur to me that IT was causing the problems I was seeing. After I got rid of that extension cable, the QH worked just fine. I have a USB2.0 compliant extension cable in the house that I'll swap it out with and see if it improves. I'm still waiting for the USB2.0 cable to connect to the QH from Craig. When it comes in, I'll install both and hopefully that'll be the end of QH com errors for anybody.

[area302]

Here is the Radio Shack link to the ferrite I used:

http://www.radioshack.com/product/index ... Id=2103222

No worries with it being to powerful from that junky store, the first one didn't even work it was so weak. (it was a small single pass ferrite bead)
No specs on it I could find.
I know it was just a crap shoot, but I lucked out and it works great.
Just remember to loop it through twice, not just a single pass, and this choke has the room for it.
I was thinking about to much suppression, but this setup has done the trick.
Hoping Craig will take care of the technical aspects of the new cable/choke he is shipping.

[Cougar5.0]

Boy, is it irritating that links open in the same tab - I keep wondering where EECTuning went after clicking a link

I love the "Tech specs" of the Rack-shack site - basically they tell you that "it's gray"

I used to occasionally use ferrite beads on the output pins of the board-mount power supplies I designed - they did indeed help with the high-frequency noise. Depending on the size, material & the input capacitance/impedance of the QH, the bead could round the transitions off a bit - easily enough seen with a 'scope. Trying to quantify the effect using specs like "it's gray" may prove to be a fools errand however

[Dale McPeters]

Well I tried a few more things last night and I am having some of the same issues that I did when I used to run my old eec-tuner board.......

I am still having drop out issues with no extra electronics hooked up in the car but at least now I get about 2 min. of data logging with BE before it looses connection and dies....hopefully that will be resolved with the new cables etc. that Craig is shipping out. I had bought another usb 2.0 cable to test with last night that was supposed to be shielded. It even said so on the package but it was not so back to the store it goes....of course it was from our friendly radio shack....

I can just about kill the data logging on the QH completely just by plugging some of my electronics back up in the car, such as MSD etc. etc. This is with both software packages that I am testing. And before someone says anything I already have an MSD 8830 cap (noise filter) wired in to the msd box. This car does have an after market coil and racing wires etc. etc.

Now I just need to get another one and wire it into the EEC side of the harness like we used to for the eec-tuner...

Take note that I do not have these problems with the AEM installed in the car so either my eec-iv's are getting old like me or I am so old that I do not remember how to do this anymore.....which is a very good possibility.
May have to buy another refurb eec just to see....but I do not have problems with the moates j3 module plugged in..

This testing is being done on a '93 cobra X3Z using an Leveno/IBM T60 2.4 dual core 2 gig ram 120 gig hard drive XP SP3.
I have also uninstalled all related software and divers and reinstalled eveything and updated XP with all available drivers etc.. Copied drivers from i386 to system 32 etc. etc.

Also started with fresh binaries for both software packages pulled straight from the eec that the testing is being done with etc. etc.

[area302]

Dale, you can do what Craig suggested to me, and just try and data log with the key on engine off, the TPS voltage.
I was able to datalog that forever till I turned the car on.
Thats when I double looped a ferrite close to the eec on the USB cable.
If it will not work with KOEO then that's another story.
I had actually given up on the first Radio Shack ferrite that didn't work, I was just using my camera USB cable with ferrite on it.
The next day I picked the cable out of the trash and tried it with the larger ferrite I had purchased(link in post above) and it worked fine.

[Dale McPeters]

Dale, you can do what Craig suggested to me, and just try and data log with the key on engine off, the TPS voltage.
I was able to datalog that forever till I turned the car on.
Thats when I double looped a ferrite close to the eec on the USB cable.
If it will not work with KOEO then that's another story.
I had actually given up on the first Radio Shack ferrite that didn't work, I was just using my camera USB cable with ferrite on it.
The next day I picked the cable out of the trash and tried it with the larger ferrite I had purchased(link in post above) and it worked fine.

I have been able to datalog the tps with the key on engine off in Paul's software for as long as I would let it run. I even had all the patch code working (except rpm) on one of the older builds. I am going to re-try the old one again tonight.

It was BE that was giving me fits even to get a full 1 min of data logging time. I am now up to 2 full min. with BE.
I have even yanked the blue tooth card on advice from Leveno/IBM tech support...

I also just got back from the electronics store. I bought another usb cable which is a high speed shielded usb 2.0 cable and an assortment of ferrites & choke coils to test. I will get it working.

[area302]

That should do it, just make sure to double loop the wire through the ferrite choke, a single pass did not work for me.

[Dale McPeters]

Ok I am up and completely running now. I installed another msd-8830 cap (noise filter) just before the eec-iv like we used to with the eec-tuner.

I also bought a new shielded high speed usb 2.0 cable, ferrits & chokes and installed one of each on the new usb cable.

My car really puts out the EMI !!!!

Note that it is running all of the electronics. I hooked everything back up in the car except for my FJO-WB as I am waiting on an NTK sensor for it.

I will do more testing when I can get the car out on the road. I am not going to today as the roads are icy and it is 29 deg. Right now…..

Both software packages are up and running also.....

-Dale

[86GT]

That is really good news. Keep us update.

[Cantedvalve]

Just got my QH. It came with a 2 naked cables and 2 ferites. So, is the answer to the dropping connection problem installing the ferits on the cables, making sure they are OUTSIDE of the EEC case?

[Dale McPeters]

Just got my QH. It came with a 2 naked cables and 2 ferites. So, is the answer to the dropping connection problem installing the ferits on the cables, making sure they are OUTSIDE of the EEC case?

Yes outside the EEC case. One close to the EEC and one close to the laptop usb connection.

Note that you may only need one so try the one at the EEC first.

[cgrey8]

The MOST important thing was getting the USB2.0 compliant cables. The originals were not. The ferrite beads are just added insurance. But unfortunately I don't think the ferrite beads are that useful at blocking noise on USB 2.0 communications.

[tealstang67]

Was craig going to send out cables? I have had some weird issues with the tunes getting corrupt. I wonder if it is the cable.

[cgrey8]

The tune shouldn't be getting corrupt. Unlike with the TwEECer, each write attempt to the QH includes a checksum, and if the checksum doesn't add up, then the QH denies the write and tells BE the checksum was bad and to resend the last write attempt. So even if the cord was causing noise, I think the protocol sufficiently protects against that.

As for getting the cables, contact Craig for details on that.

[SigEpBlue]

I'm beginning to wonder whether there is, or should be, some form of isolation, optical or otherwise, to help with this issue.

[Dale McPeters]

The MOST important thing was getting the USB2.0 compliant cables. The originals were not. The ferrite beads are just added insurance. But unfortunately I don't think the ferrite beads are that useful at blocking noise on USB 2.0 communications.

I had a usb 2.0 cable that came with the test QH that Craig sent but it was not shielded. So I bought a usb 2.0 shielded high speed cable and still had to install the ferrites to get it to work reliably in my '93 cobra.......so there....

[cgrey8]

Hmm...interesting. I didn't realize the ferrite beads made that much difference and benefit. That's encouraging.

[Cantedvalve]

So the cable that came with the kit is not USB 2.0? I dont remember markings, but I will check it when I get home tonight.

[Cougar5.0]

I don't have the QH, but the USB cable I'm using has a nice ferrite bead that looks like it's molded right into the cable.

[ys5259]

Hi all, New to the forum and tuning for that matter but thought my $0.02 might have some value here.

I just purchased a QH a couple of weeks ago along with the BE and EA software. I ran into the same datalogging problem as described in this post where my symptoms where that with vehicle at standstill no problems with datalogging but when driving the datalog operation would just stop and could not restart without disconnecting and reconnect USB and restarting BE. I could however log as long as I wanted at a standstill. The first things I checked were all my connections as it seemed that the common factor for the disonnection was the motion of the vehicle and something was loose but could not find anything. My kit did come with the 2 piece cable and ferrite beads. Asked for help at Moates and was given advice on ferrite placement and told that ignition interference could cause some connectivity problems. I'm using a hot coil and multispark ignition system so before I disconnected these to test this idea I came here to see if there were similar encounters with similar equipment and maybe a remedy.

To make a long story short I tried the suggestion posted by Gearhead using the Radio Shack ferrite with a double pass right after coming out of the EEC and this seems to have solved the problem. I have been able datalog while driving without a drop out for 5 days now. I tried this first because it was cheap, quick, and easy so worth a shot. To be honest I didn't really think it would work. I also argee with the comment about it being difficult to quantify the problem/solution without any real specifications of the ferrite bead but thought it important to note another case where this has solved this issue as it may benefit others.