Skip to content
Projects
Groups
Snippets
Help
Loading...
Sign in
Toggle navigation
C
Conv TTL Blocking - Hardware
Project
Project
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
Wiki
Wiki
image/svg+xml
Discourse
Discourse
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Commits
Issue Boards
Open sidebar
Projects
Conv TTL Blocking - Hardware
Commits
91a25598
Commit
91a25598
authored
Mar 10, 2017
by
Denia Bouhired-Ferrag
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
Doc: updated guide with limit on number of blocking channels per crate.
parent
00f551e6
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
46 additions
and
10 deletions
+46
-10
hwg-conv-ttl-blo.bib
doc/hwg/hwg-conv-ttl-blo.bib
+1
-1
hwg-conv-ttl-blo.tex
doc/hwg/hwg-conv-ttl-blo.tex
+45
-9
No files found.
doc/hwg/hwg-conv-ttl-blo.bib
View file @
91a25598
...
...
@@ -42,7 +42,7 @@ Date = {July 2015}
@misc{sch,
title = {{CONV-TTL-BLO Schematics}},
howpublished = {\url{https://edms.cern.ch/file/1278535/1/EDA-02446-V
2-1
_sch.pdf}}
howpublished = {\url{https://edms.cern.ch/file/1278535/1/EDA-02446-V
4
_sch.pdf}}
}
@misc{rtmm-sch,
...
...
doc/hwg/hwg-conv-ttl-blo.tex
View file @
91a25598
...
...
@@ -524,13 +524,6 @@ For more details about this common-impedance coupling event, see~\cite{crosstalk
\vspace
{
11pt
}
\noindent
\textbf
{
\textit
{
Note that if the FPGA is improperly configured, a DC high-level signal
on the power MOSFET's grid pin will yield a too high current passing through the MOSFET,
which will lead to its failure. Make sure the FPGA pins driving the blocking output stage
are properly configured to drive time-limited pulses or a DC low-level at the power
MOSFET's grid, or that the blocking output enable signal from the FPGA is low, so as to
keep the outputs of the tri-state buffers in high-impedance.
}}
%------------------------------------------------------------------------------
% SEC: Blocking pulse rep
%------------------------------------------------------------------------------
...
...
@@ -623,7 +616,7 @@ for buffering the VME signals. The control and data lines of the chip are driven
by logic within the FPGA, which controls lighting of each of the LEDs. An example
of how the LEDs can be driven using the FPGA is given in Section~5 of~
\cite
{
ctb-hdlguide
}
.
TTL (front panel), INV channels and blocking (rear panel) pulse LEDs are driven by the FPGA
TTL (front panel), INV channels
(front panel)
and blocking (rear panel) pulse LEDs are driven by the FPGA
via a SN7414 Schmitt trigger. In the case of the blocking LEDs, the output of
the Schmitt trigger is connected directly to the VME P2 connector and through
the RTM to the piggyback, where the current-limiting resistor and the LEDare located.
...
...
@@ -734,7 +727,7 @@ calculations below.
\subsection
{
Minimum blocking pulse level
}
\label
{
app:blo-min-level
}
The optocoupler LED has a
forward voltage of 1.5
~V and therefore when the LED is on,
The optocoupler LED has a
maximal forward voltage of 1.8
~V and therefore when the LED is on,
the voltage in point 2 of Figure~
\ref
{
fig:blo-inp
}
is
\begin{equation}
...
...
@@ -827,6 +820,24 @@ I_{LED,RMS} = I_{LED} \sqrt{\delta}
\section
{
Blocking output stage calculations
}
\label
{
app:blo-outp
}
\textbf
{
\textit
{
Since the following calculations were originally intended to document design
decisions
related with v2.1 boards and implemented in release 3.0 of the gateware, they are no longer
applicable for v4 boards.
\\
Indeed, the blocking output protection discussed in Sec.
\ref
{
sec:blo-outp
}
and in
\cite
{
blo-out-protect
}
, means that frequency of repetition can be considerably
increased, by relying on adequate protection in hardware (current and pulse width limitation) and
in the gateware (Implementation of an thermal model of the
blocking output circuit inside the FPGA, effectively resulting in dynamic regulation of output
frequency).
\\
Nonetheless these calculations are still relevant to v2
.1 boards and earlier, and the final recommended duty cycle will still be applicable in these
boards,
even with future releases of the gateware. Moreover, the calculations presented here remain of
great value and demonstrate the procedure to follow to translate MOSFET datasheet thermal
characteristics, into switching frequency limits similar to the calculations in
\ref
{
sec:blo-outp
}
and
\cite
{
an10273
}
.
\\
}}
\vspace*
{
11pt
}
Figure~
\ref
{
fig:blo-outp
}
shows the blocking output stage, as a reference for the
calculations below.
...
...
@@ -1007,6 +1018,31 @@ T_{min, final} = 241 {\mu}s
f
_{
max, final
}
=
\frac
{
1
}{
T
_{
min, final
}}
= 4150 Hz
\end{equation}
%==============================================================================
% APP: Block outp stage calc in the burst mode
%==============================================================================
\pagebreak
\section
{
Frequency limits implemented in Version 4 boards
}
\label
{
app:freq-limits
}
From experimental measurements and functional specifications, a 2MHz maximum frequency limit is
implemented in gateware for
$
250
ns
$
-wide pulses and 104kHz for
$
1
.
2
\mu
s
$
~
\cite
{
ctb-ug
}
. After
tests on the v3.0 prototypes it was found that jitter in the pulses at the blocking input of the
FPGA, mean that pulse repetition at maximal frequency (2MHz) corresponding to periods of
$
500
ns
$
,
is difficult to achieve with any consistency. In effect it means that some pulses, may be missed if
jitter causes them to be shorter than the required
$
500
ns
$
. As a result, internal to the
FPGA, a lower pulse limit is implemented, exactly one clock cycle shorter (One system clock cycle
corresponds to
$
50
ns
$
), to allow for errors caused through jitter.
\\
The gateware limits therefore set maximum repetition frequency for
$
250
ns
$
-wide pulses at ~2.22
MHz, corresponding to
$
450
ns
$
. The limit is~
$
\approx
104
.
7
kHz
$
, i.e.
$
9550
ns
$
instead of the
required
$
9600
ns
$
for wide
$
1
.
2
\mu
$
s pulses.
\\
The guaranteed, error-free, repetition times are therefore as advertised in the user guide
\cite
{
ctb-ug
}
,
$
500
ns
$
and
$
9600
ns
$
for the short and wide modes of operation.
\\
\textbf
{
\textit
{
Repetition at rates below these values is possible, but is not
guaranteed to be error-free, and therefore not advertised in the user manual, nor is it recommended.
}}
%==============================================================================
\end{appendices}
%==============================================================================
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment