Re: oscillations

[Gerard Visser <gvisser@indiana.edu>]

Hall D Electronics:

Hi Fernando,
--->
	Gerard

Fernando J. Barbosa wrote:
> Hi Gerard and Yves,
> 
...>
> I also see great signals from Yves' log.

Yes! (And sorry Yves' I haven't written back about that yet...)

  But what exactly caused the
> oscillations in the first place? I have not seen any good explanations
> as for the cause so far but I am happy for the progress. Perhaps you and
> Yves could right a note on the findings.

I will try to reproduce it here in my lab and give some quantitative 
data. But, here are my observations (in sequence) regarding the 
oscillation, on first day at CMU:

1. Chamber was running with CLAS preamps. I looked at signals, and also 
noted the severe interference. Scope FFT indicated 88.3 MHz and this 
supports the theory that main problem was FM radio, apparently there is 
a CMU-affiliated transmitter somewhere within 1/2 mile, possibly much 
closer than that. [Curtis/Yves I still think maybe you should 
investigate exactly where that antenna is. Maybe it's on your building's 
roof!]

2. We put on the new preamp and I asked Yves to connect as before. 
Oscillations were observed. Here are the details of the connection: The 
signal cable was ~3m, ERNI connector on one end, the other end broken 
out to 100 (?) Ohm termination resistors on pair #1, 2-9, and 18-25. 
Pairs 10-17 were not terminated. (They are also open on the preamp board 
  end of the cable, since no ASIC or termination resistor is there.) For 
one of the channels (at a time, we looked at a few) the 100 Ohm resistor 
termination was removed (Yves, is that right, or did we leave it?) and 
in it's place two RG-174 cables of probably unequal length, LEMO to clip 
leads, were connected. Shield was tied to shield, one center conductor 
to one side of the signal pair and the other center conductor to the 
other side of the signal pair. The wire/clip lead lengths were on the 
order of 10-20 inches total, and nothing was twisted, i.e., basically it 
was a circular loop of circumference 10-20 inches. These two RG-174 
cables connected directly (through LEMO-BNC adapters) to two channels of 
the scope, set to 50 Ohm termination.

3. I remarked that this may be a sub-optimal way of probing the signals. 
We substituted my matching pad / transformer widget (see fig. 19 in 
GlueX-doc-955-v1 - the center tap was _not_ grounded here) for the pair 
of RG-174 cables with clip leads. No oscillations were observed. We 
moved on to other stuff (attempts to tackle the EMI issue).

I should also say, I have _never_ seen any evidence of instability of 
the new preamp board under any reasonable operating conditions, so I 
don't think this is something we need to focus much effort on.

There is also the possibility that one or more of the LEMO connectors or 
cables or adapters had a flaky ground connection. Certainly we found one 
or two such things later, you'll not be surprised.

> 
> Yves - you may want to try turning off any fluorescent bulbs you have in
> the lab for two reasons: the ballast transformers, if present, can be
> very noisy; and the UV from the bulbs may cause extraneous pulses on the
> straws through the gaps from the wrapping. I first observed this at BNL
> (E850) where we had aluminized mylar straws. The UV from the bulbs
> overhead, and which penetrated the straws through the gaps in the
> wrapping, would knock electrons out of the aluminum surface inside the
> straws and produce very noticeable pulses. 

This is a very good point! At the moment I think the whole chamber is 
wrapped in aluminum foil, that takes care of the light too. But care 
should be taken not to have UV light contamination on the straws. (There 
is a "thick" lexan(?) sheet over the top of the CDC prototype for 
mechanical protection, maybe that was already enough to block the UV.)

The surface work function on
> your copperized Kapton straws is much higher so that this may not be a
> problem for you. 

FYI the metallization on the kapton straws is still aluminum, and I 
don't think there's any plans to change that.

Another interesting observation was that we could make
> the inside of the straws glow blue (plasma) in the dark at higher
> voltages, in the streamer regime. The blue color was perhaps because of
> the Argon (I believe we used Ar-ethane 80/20).
> 
> Thanks for the feedback. Regards,
> Fernando