Looper's Delight Archive Top (Search)
Date Index
Thread Index
Author Index
Looper's Delight Home
Mailing List Info

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

RE: EDP and a BIG burst of static/hiss



Sorry for the long rambling post, but I figured I'd better answer some of
the recent EDP complaints: -

It would appear that some of the memory in the last 200 or so EDPs that
Trace Elliot built has a problem with bad contacts. This was the same 
memory
bought from MemoryX in the US that was used in the previous 600 or so 
units.
It hadn't given much trouble at all for a couple of years so we figured it
was a safe bet; very few SIMMs actually fail at any of the test stages. We
also hadn't had any reports of failure in the field. However, I don't want
to spread panic; current failure figures are still a very small percentage
of the total number of SIMMs used. Still unacceptable, I know and we are
investigating the problem.
I am looking at memory from different sources at the moment but it is a
difficult commodity to get in quantity due to it being pretty much 
obsolete.
There are a few suppliers who make new stock, but they charge up to $20 per
SIMM. Many just supply re-furbished sticks pulled from old computer 
hardware
and you have no idea of the history. We have to find good quality new stock
for $4/SIMM to keep the current retail price. Any suggestions for suppliers
would be appreciated, but I'm sure I've tried most. Most suppliers laugh
when I give them the spec for the memory: You want what? How many? Good 
luck
mate. We currently need about 3000 pieces and I've got various samples on
the way to test.

'Gunk' around some of the components, mainly the power supply caps, was
added to reduce the risk of the component legs breaking through their 
solder
connection due to vibration. I had a couple of units back where the large
capacitor had a cracked joint and the unit kept re-booting as the leg 
warmed
up, cooled down, or moved through any vibration. One of these lived in a
rack on top of a bass stack so was subjected to massive vibration for hours
at a time. The hot-melt glue around these sensitive components now has
definitely helped this problem.

Our policy when I worked at Trace was immediate swap out if a faulty unit
was within a year of purchase and then 48hour turnaround on units outside
the first year. Having spent many years as the Service Manager, I helped
develop the company's repair/exchange policy and we had a very good
reputation because of it. The philosophy that we instilled in the service
engineers was that every fault was an opportunity. A customer who never has
a problem may promote your gear by playing it or quietly singing it's
praises when asked but an irate customer with a fault, (and boy did I have
some screamers!), can be turned into a friend of the company for life.
Invite him down, explain exactly what you're doing at every step, take him
out to the pub for a beer and give him a whole bag of swag when he leaves.
If it was a construction problem, report the fault to the production floor
immediately so that it doesn't happen again and if it's a design issue, go
see the R&D guys. I'm still in contact with bass players whose amps I fixed
15 years ago and this makes me feel that some of the things we did were
really worthwhile. This type of policy does cost money, (some of our amps
were $3000 a throw), but I hope it didn't contribute too much to our
eventual downfall, (that may have had more to do with uncle Henry:)). To 
get
to the point, Shane does a very difficult job; he's covering the whole US
pretty much on his own, has a product base of 25 years Trace Bass & Trace
Acoustic manufacture to look after and rarely has new units in stock to 
swap
out for faulty ones. Clearly Gibson needs to look at the requirements and
procedures of their EDP back up and I hope that things will change when we
build large quantities of the new version. They have never been as keen to
promote the EDP as they now have with Kevin Van Pamel at the controls and
things can only improve. I can only suggest things to them, as I am not an
employee anymore but he does genuinely listen. For diplomatic reasons I
won't comment anymore on policy at Gibson. It's a shame you guys with
problems are not in the UK; it would be a pleasure to meet you, swap your
units out or fix them for you over a stout or two, (I actually prefer
Belgium wheat beer for future reference). That's one benefit of being in
this a very small and often wet nation; where Trace was situated, you could
drive from pretty much anywhere in the country in a few hours to see us.

On to the quality control issue. The sequence of events that each Trace
Elliot built unit went through is as follows: -
PCB and parts assembled in Tunisia.
Completed PCB inspected for dry/bad joints and then tested with a
semi-automated test jig and software written by Matthias.
Board shipped to the UK, inspected and then re-tested with an identical 
test
jig.
Unit assembled and placed on a test rack where all foot control functions
are tested and noise levels checked with a TA100 reference amplifier.
Multiple short loops are entered with overdub by a guitarist with F/B on
100%
Unit gently banged on all sides while running to check for loose internal
connections.
Unit left to run the loops for 24 hours and then carefully listened to by
same guitarist to check for any addition noise, loop decay or distortion.
Unit packed & shipped.

All future Straight Edge production of the Echoplex will be in the UK but 
we
were very happy with the quality from Tunisia when Trace built them.

Some units can slip through the net for obscure reasons; I had few 
instances
of sales guys sneaking into the factory when the staff had gone for the day
and taking un-tested stock out but not any EDPs as far as I know. However,
this type of thing is very rare and so each product that left the factory
had been through some pretty rigorous testing. Much more in fact that many
other audio manufacturers do and I've seen a lot of production lines in my
time.
I'm not trying to make excuses but I'm confident that every EDP that we 
made
was working perfectly when it left. The bent pin on the EPROM is a bad
mistake on the part of two test engineers and I apologise on their behalf.
One of them now works for me at Straight Edge so I've already mentioned it
to him and he promises to be more vigilant in the future, (it may have been
the other guy). If anyone wants to suggest any further testing before
despatch, I'd be happy to consider it.

I hope some of this helps, (if you've read down this far). We always 
welcome
constructive criticism or ways to improve things.
Andy.


-----Original Message-----
From: sine@zerocrossing.net [mailto:sine@zerocrossing.net]
Sent: 10 September 2002 19:13
To: Loopers-Delight@loopers-delight.com
Subject: Re: EDP and a BIG burst of static/hiss

Hey kids.

Again, I want to say thanks for all the info and help.  I didn't know 
enough
about the EPROM burning process to realize that it was probably not
software.

So, what I did last night was take the whole thing apart (again) and clean
that baby up.  I took each EPROM and memory chip (which I had not 
previously
touched, upon finding the bent EPROM pin... yes Kim, it was in fact in 
there
after I fixed it) and scrubbed it all down with contact cleaner.  The 
EPROMs
seemed fine, but I did notice a fair amount of black on the cuetip after
cleaning off the memory chips.  I then put it all back together and, of
course, it worked fine.  Played with it for a few hours with no problem and
went to bed with a loop running and woke up with it still there.

I wouldn't call this a good test, but it was barely staying booted up for 5
minutes before I did the big clean.  I'm going to try and repeat this every
night to make sure everything's good.

I still feel this is kind of weird though.  I've got a Mac SE30 that still
runs with no problems.  I also had a Mac IIX that ran for years and years
and
years.  In the hot humid summers and bitter cold winters of upstate NY.
I've
never had to clean a contact or reseat anything, yet this seems common with
the EDP.  Someone said it was because of the fact we take the EDP out and
bang
it around a lot more than computer gear, but I used to take the SE30 around
with me to gigs as a MIDI sequencer, so it often went to gigs in sub 0
conditions, only to find itself in a heated enviornment moments later.  The
thing still boots up.  I still use it to run the "Fish" screensaver in leu
of
a real fishtank.

Could it be the type/brand of ram Gibson uses?  Why not clean it with a
cleaner that protects it from oxididation before installing it?  My unit 
was
a
newer Gibson unit that's probably been a San Rafael CA storeage room most 
of
it's life.  That can't be a bad climate for electronics gear.  Moderate and
dry all year.  I know shit happens, but something tells me that other 
forces
are at work in this case.  We need to tell those Brits to stay away from 
the
stout and start QC'n their gear before they ship it.

Mark Sottilaro

Kim Flint wrote:

> At 03:09 PM 9/9/2002, sine@zerocrossing.net wrote:
> >First, I want to thank you all for your excellent role as "Gibson
customer
> >support" and for all the good information.  Tonight I shall (once again)
> >open up my EDP and this time attempt to clean all the contacts.  I only
> >reseated the eproms, and upon finding the bent pin, I figured that was
> >obviously my problem.  When it worked after that, I thought I was done.
I
> >didn't even bother with the memory becaues the fault seemed so obvious.
>
> You might also want to recheck the rom pins too. On occasion I've bent a
> rom pin back, but didn't do it very well and left a kink in it. Then when
I
> pressed it into the socket I just bent it out all over again. It's worth
it
> to take some care and make sure the are all really in there.
>
> kim
>
> ______________________________________________________________________
> Kim Flint                     | Looper's Delight
> kflint@loopers-delight.com    | http://www.loopers-delight.com