1: % last update:
2: %
3: % Questions:
4: \documentclass[12pt,preprint]{aastex}
5: %
6: %\usepackage{natbib}
7:
8:
9: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
10: \newcommand{\PSbox}[3]{\mbox{\rule{0in}{#3}\special{psfile=#1}\hspace{#2}}}
11: \newcommand{\FigNum}[1]{\unitlength 1pt \begin{picture}(55,10)(-400,35)
12: \put(0,0){Figure #1}
13: \end{picture}}
14: \newcommand{\perval}[2]{{#1\mbox{$^{#2}$}}}
15: \newcommand{\msun}{\mbox{$M_\odot$}}
16: \newcommand{\rsun}{$R_\odot$}
17: \newcommand{\persec}{\perval{\rm s}{-1}\/}
18: \newcommand{\percm}{\mbox{$\cm^{-2}$}}
19: \newcommand{\peryear}{{{\rm yr}$^{-1}$}}
20: \newcommand{\ppm}{\mbox{$\pm$}}
21: \newcommand{\cgsflux}{\erg\,\percm\,\persec}
22: \newcommand{\photflux}{{\rm phot}\,\percm\,\persec}
23: \newcommand{\cgslum}{\erg~\persec}
24: \newcommand{\kmpersec}{\km~\persec}
25: \newcommand{\approxlt}{\mbox{$\lesssim$}}
26: \newcommand{\approxgt}{\mbox{$\gtrsim$}}
27: \def\etal{{et~al.}}
28: \def\eg{{e.g.}}
29: \def\ie{{i.e.}}
30: \def\ebv{{$E(B-V)$}}
31: \newcommand{\nh}{\mbox{$N_{\rm H}$}}
32: \newcommand{\nhtt}{\mbox{$N_{\rm H, 22}$}}
33: \def\chisqr{\mbox{$\chi^2$}}
34: \def\chisqrnu{\mbox{$\chi^2_\nu$}}
35: \newcommand{\ud}[2]{\mbox{$^{+ #1}_{- #2}$}}
36: \newcommand{\supp}[1]{\mbox{~~$^{#1}$}}
37: \newcommand{\ee}[1]{\mbox{$10^{#1}$}}
38: \newcommand{\tee}[1]{\mbox{$\times 10^{#1}$}}
39: \newcommand{\tento}[1]{\mbox{$10^{#1}$}}
40: \newcommand{\keV}{\mbox{$\rm\,keV$}}
41: \newcommand{\MeV}{\mbox{$\rm\,MeV$}}
42: \newcommand{\cm}{\mbox{$\rm\,cm$}}
43: \newcommand{\km}{\mbox{$\rm\,km$}}
44: \newcommand{\second}{\mbox{$\rm\,s$}}
45: \newcommand{\ksec}{\mbox{$\rm\,ksec$}}
46: \newcommand{\yr}{\mbox{$\rm\,yr$}}
47: \newcommand{\GramPerCc}{\mbox{$\rm\,g\,cm^{-3}$}}
48: \newcommand{\K}{\mbox{$\rm\,K$}}
49: \newcommand{\erg}{\mbox{$\rm\,erg$}\/}
50: \newcommand{\gauss}{\mbox{$\rm\,G$}}
51: \newcommand{\Qnuc}{Q_{\rm nuc}}
52: \newcommand{\Msun}{\mbox{$\,M_\odot$}}
53: \newcommand{\mJy}{\mbox{$\rm\,mJy$}}
54: \newcommand{\uJy}{\mbox{$\rm\,\mu Jy$}}
55: \newcommand{\Jy}{\mbox{$\rm\, Jy$}}
56: \newcommand{\axaf}{{\em Chandra\/}}
57: \newcommand{\chandra}{{\em Chandra\/}}
58: \newcommand{\rosat}{{\em ROSAT\/}}
59: \newcommand{\asca}{{\em ASCA\/}}
60: \newcommand{\rxte}{{\em RXTE\/}}
61: \newcommand{\xmm}{{\em XMM\/}}
62: \newcommand{\beppo}{{\em BeppoSAX\/}}
63: \newcommand{\sax}{{\em BeppoSAX\/}}
64: \newcommand{\swift}{{\em SWIFT\/}}
65: \newcommand{\kpc}{\mbox{$\rm\,kpc$}}
66: \newcommand{\Aem}{\mbox{$A_{\rm em}$}}
67: \newcommand{\Teff}{T_{\rm eff}}
68: \newcommand{\kteff}{$kT_{\rm eff}$}
69: \newcommand{\lp}{{LP~944-20}}
70: \newcommand{\usno}{{USNO-A2}}
71: \newcommand{\hal}{{H$\alpha$}}
72: \newcommand{\sig}{{$\sigma$}}
73: \newcommand\lsun{L$_\odot$} % solar luminosity
74: \newcommand\lxlbol{$L_{X}$/$L_{\rm bol}$}
75: \newcommand\lx{$L_{X}$}
76: \newcommand\psr{PSR~B1821$-$24}
77: \newcommand\musec{$\mu$s}
78: \newcommand\msec{\mbox{${\rm ms}$}}
79: \newcommand\ms{{\msec}}
80: \newcommand\nancay{Nan\c cay}
81: \newcommand{\psrp}{3.05~ms}
82: \newcommand{\nus}{\mbox{$\nu_{\rm s}$}}
83: \newcommand{\dmunit}{pc~\perval{cm}{-3}}
84: \newcommand{\cperksec}{c~\perval{ksec}{-1}}
85: \newcommand{\fxfr}{\mbox{$F_{X}/F_{\rm 1.4\,GHz}$}}
86: \newcommand{\fxlim}{\mbox{$F_{X, {\rm lim}}$}}
87: \def\cxo{CXOU J191121.3+003844} % NOTE: IAU standard is to truncate
88: % celestial coords to make the name.
89:
90: \def\rrat{RRAT~J1819$-$1458}
91: \def\aql{Aql~X-1}
92: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%the document begins%%%%%%
93: %\clubpenalty=10000
94: %\widowpenalty=10000
95: \received{}
96: \revised{}
97: % Manuscript #
98: %\accepted{}
99: \slugcomment{}
100:
101: \begin{document}
102:
103: \title{Constraints on RRAT Emission Mechanisms from {\em RXTE}/PCA
104: Observations of \rrat}
105:
106: \author{Robert E. Rutledge\altaffilmark{1}
107: }
108:
109: \altaffiltext{1}{Department of Physics, McGill University, 3600 rue
110: University, Montreal, QC, H3A 2T8, Canada; rutledge@physics.mcgill.ca}
111:
112: \begin{abstract}
113: We derive the second and most stringent limit to date of the
114: X-ray/radio flux ratio (\fxfr) for the radio bursts associated with
115: the recently identified source class, the Rotating Radio Transients
116: (RRATs). We analyze 20.1 hr of \rxte/PCA observations of \rrat\ -- a
117: period during which 350\ppm23 RRAT radio bursts occurred, based on the
118: previously observed average radio burst rate. No X-ray bursts were
119: detected, implying an upper-limit on the X-ray flux for RRAT-bursts of
120: $\fxlim<$1.5\tee{-8} \cgsflux\ (2-10\,\keV), or a luminosity
121: $<2.3\tee{37} (d/3.6\kpc)^2$\cgslum. The time-average burst flux is
122: $<$2\tee{-13} \cgsflux\ (0.5-8 keV) -- a factor of 10 below that of
123: the previously identified persistent X-ray counterpart. Thus, X-ray
124: bursts from the RRAT are energetically unimportant compared with the
125: persistent X-ray emission. From the previously observed burst radio
126: flux, we derive an upper-limit $\fxfr\leq4.2\tee{-12}
127: \cgsflux\perval{\rm mJy}{-1}$ for the radio bursts from this RRAT, the
128: most stringent to date, due to the high radio flux of bursts from this
129: source. The \fxfr\ ratio is a factor $\sim$80 larger than that of the
130: millisecond pulsar PSR~B1821-24; thus emission processes of
131: X-ray/radio efficiency comparable to MSP pulses cannot be ruled out.
132: However, if the RRAT burst emission mechanism is identical to the
133: \msec\ bursts of magnetars, then the \msec\ bursts of magnetars should
134: be easily detected with radio instrumentation; yet none have been
135: reported to date.
136: \end{abstract}
137:
138: \keywords{stars: flare; stars: neutron; X-rays: stars; pulsars:
139: individual (J1819$-$1458)}
140:
141: \section{Introduction}
142:
143: Recently, a new observational class of neutron stars has been proposed
144: (\citealt{mclaughlin06}, M06 hereafter), Comprised of eleven sources, the class is
145: defined by the phenomenon of fast (2-30 ms) radio bursts, which repeat
146: on timescales of 4\,min - 3\,hr. Periodicities were claimed to be
147: found in the range of $P=$0.4-7\,s for ten of the eleven sources,
148: through period folding; this suggested association with rotating
149: neutron stars. In three of the eleven sources, period derivatives
150: were claimed, with one suggesting a magnetic field strength of
151: 5\tee{13} G -- a possible magnetar. Because the $P-\dot{P}$ analysis
152: remains unpublished and undescribed, we regard association of the
153: radio-bursting population with neutron stars to be an unsupported
154: hypothesis; nonetheless, we adopt the hypothesis and nomenclature of
155: referring to the sources as Rotating RAdio Transients (RRATs), and
156: consider their association with neutron stars to be the leading
157: hypothesis.
158:
159: One of these -- \rrat -- has a $7\arcsec\ \times32\arcsec$ error
160: ellipse (M06)\footnote{It is unclear why the positional uncertainty
161: from this source is so much smaller than, for example, the
162: $15\arcmin\times7\arcmin$ error region for RRAT J1911+00, obtained
163: in the same study with the same instrumentation.}, which permitted discovery of an X-ray counterpart, CXOU
164: J181934.1$-$145804 \citep{reynolds06}, with absorption-corrected flux
165: of 2\tee{-12} \cgsflux\ (0.5-8 keV), or 3.1\tee{33} \cgslum\ at the
166: $d=3.6$\,\kpc\ dispersion-measure-indicated distance. Two-hundred
167: twenty nine radio bursts at $\nu=$1.4 GHz, with average half-duration
168: $w_{50}$=3\,\msec\ were observed in 13 hours of radio integration, the
169: brightest with a peak flux of $S_\nu=3.6\,\Jy$. If we take $w_{50}$
170: to be the light crossing time of the emission region, this implies a
171: source brightness temperature (e.g., \citealt{guedel02}):
172: % ARAA 2002
173:
174: \begin{equation}
175: T_b = 3\times10^{23} \left(\frac{S_\nu}{3.6\, {\rm Jy}}\right)
176: \left( \frac{ 1.4\, {\rm GHz}}{\nu}\right)^2
177: \left( \frac{3\,{\rm ms}}{w_{\rm 50}} \right)^2
178: \left( \frac{d}{3.6\, {\rm kpc}} \right)^2 \, {\rm K }
179: \end{equation}
180:
181: \noindent This is significantly above the brightness temperature
182: limit for inverse Compton scattering ($T_B\sim\ee{12} K$)
183: % Kellermann & Pauliny-togh 1969
184: which likely requires a coherent emission mechanism for this and
185: other RRATs, possibly accompanied by X-ray emission.
186:
187: In this paper, we analyze \rxte/PCA data, and search for X-ray bursts
188: which may accompany the observed radio bursts, finding none. In
189: \S~\ref{sec:obs}, we describe the observations and analysis; in
190: \S~\ref{sec:results}, we present the results of the X-ray burst
191: search; and in \S~\ref{sec:con}, we discuss the implications of these
192: results and conclude.
193:
194: \section{Observations and Analysis}
195: \label{sec:obs}
196:
197: We analyze observations taken in \rxte/PCA \citep{pca,pca2}, obtained
198: from the public archive (Table~\ref{tab:obs}). For basic data
199: extraction, we used
200: FTOOLS\footnote{http://heasarc.gsfc.nasa.gov/ftools} \citep{ftools}
201: following standard \rxte\ recipes. We used {\tt GoodXenon\_16s}-type
202: data, analyzing 20.1 hr of data. We performed some data selections:
203: we excluded data from PCU1 (Proportional Counter Unit 1, numbered
204: 0-4), and layers 2-3 from all PCUs, and required at least 2 active
205: PCUs. Most observations subsequently had 3 or 4 active PCUs (0,2-4; or
206: 0,2-3). These selections eliminate the PCU high-voltage
207: breakdowns\footnote{HV breakdown in PCU3 and PCU4 is typically marked
208: by a burst of counts, usually in PHA channels $<$20, and limited to a
209: single wire within the PCU, on a $\leq2\msec$ timescale. However,
210: breakdown in PCU1 is marked by low-PHA counts across {\em all} wires,
211: and is often terminated by a single high-energy (PHA 15-30) count; it
212: may be due to a distinct phenomenon.} which occur in PCU1, PCU3 and
213: PCU4 \citep{pca2} in these and other data on which the programs were
214: tested.
215:
216: We examined pulse-height-analyzer (PHA) channels only $\leq27$
217: corresponding to an approximate energy range of 1.5-10 keV, and
218: excluded propane-veto layer events.
219:
220: We searched for bursts on two timescales: $\tau=3.0$\,\msec\ (the
221: observed radio-burst timescale for \rrat) and 1/10 that value
222: (0.3\,\msec), to cover possible shorter X-ray duty cycles.
223:
224: We measure the number of counts $M$ over a period of $\pm$1 sec about
225: each count in the data stream. We measure the number $N$ of counts
226: which are observed within $\pm \tau/2$ of each count, and we calculate
227: the Poisson probability of measuring $\geq N$ counts for an expected
228: number of counts $\mu=M\times\tau/(2\, {\rm s})$ as:
229:
230: $$
231: P_{\rm Poisson}(\mu, \geq N) = \sum_{i=N}^\infty \frac{\mu^N}{N!}\exp(-\mu)
232: $$
233:
234: Summing over all 22 observations, we observed $N_{\rm phot}=3069964$
235: total counts. An event for which $P_{\rm Poisson}\times N_{\rm
236: phot}<0.01$ would have 1\% chance of being produced due to random
237: fluctuation alone during the entirety of our analysis; we call such an
238: event a ``trigger'', and store all information about the $N+1$ counts
239: in that trigger for future examination.
240:
241: \section{Results}
242: \label{sec:results}
243: We found no triggers in the above data stream, on either timescale. We
244: calculated the limits on burst rate as a function of average burst
245: flux over the 3\,ms and 0.3\,ms timescale (Fig.~\ref{fig:limits}). At
246: the time of each count, we calculated the average (over $\tau$) number
247: of counts per PCU per sec, corresponding to a 99\% confidence
248: countrate fluctuation (that is, a Poisson fluctuation which would only
249: occur, on average, in a dataset with 100$\times$ as many counts as the
250: present dataset); this resulted in 99\% confidence upper-limits
251: between 422-1010 c/s/PCU on the $\tau=3\msec$ timescale). During
252: these observations, \rrat\ lay 27.5\arcmin\ off-axis, at a collimator
253: efficiency of 0.45-0.5, for which we correct the countrates. Layer-1
254: wires in the PCUs make up $\approx$90\% of the total active area below
255: 10 keV, for which we correct the countrate (the remaining 10\% is in
256: layers 2-3, which we neglected). Finally, we use an energy correction
257: factor of 1.0\tee{-8} \cgsflux (2-10\,\keV) per 1000 c/s/PCU (assuming
258: a photon power-law slope $\alpha$=1, and neglecting \nh\footnote{using
259: WebPIMMS; http://heasarc.gsfc.nasa.gov/Tools/w3pimms.html} which we
260: assume as the default spectrum throughout), where here, the countrates
261: are for layers 1-3. The maximum burst rate is the 95\% confidence
262: upper-limit for the burst rate, $2./T(>F_{\rm lim})$, where $T(>F_{\rm
263: lim})$ is the total integrated observing time sensitive to bursts from
264: \rrat\ with flux $>F_{\rm lim}$; the numerator (2) comes from the
265: number of bursts for which one detects 0 bursts $\leq$5\% of the time.
266:
267: In general, flux sensitivity did not change much during observations;
268: thus, we detected no bursts on a $\tau=3\,\msec$ ($\tau=0.3\,\msec$)
269: timescale at fluxes above $\approxgt1.5\tee{-8} \cgsflux$ ($8\tee{-8}
270: \cgsflux$), with 99\% confidence, with burst rates as low as
271: 0.1\,\perval{hr}{-1}. At lower fluxes, the observations rapidly
272: become insensitive.
273:
274: \section{Discussion and Conclusions}
275: \label{sec:con}
276:
277: \rrat\ is the most prolific radio-burster of all the reported RRATs --
278: 229 radio bursts detected in 13\,hr of radio observations. On
279: average, we would expect 350\ppm23 such radio bursts to have occurred
280: during these observations, assuming the radio burst rates do not
281: change with time. None of these had a corresponding X-ray flux
282: $\geq$1.5\tee{-8} \cgsflux (2-10 keV). We expect 2, on average, to
283: have been as radio bright as the brightest previously observed (3.6
284: Jy), which implies that $\geq$1 such bursts were emitted (95\%
285: confidence). This, in combination with our derived burst flux limit,
286: results in an X-ray to radio flux ratio limit for bursts from \rrat\ of
287: approximately:
288:
289: $$
290: \fxfr \leq 4.2\tee{-12}\, \cgsflux\,{\rm mJy}^{-1}
291: $$
292:
293: \noindent This limit is a factor of 14 lower than that derived for
294: RRAT~J1911+00 \cite{hoffman06}, due to the factor of 14 greater radio
295: flux observed from the brightest radio burst from \rrat\
296: (M06). It may be instructive to compare this ratio
297: with that from other transient coherent emission processes.
298:
299: The limit corresponds to \fxfr$\leq\ee{14.7}$ Hz, comparable to the average
300: ratio observed from stellar coronal activity
301: \citep{guedel93}; however the timescales for coronal flares (for
302: example) are typically $\sim10^3\,\sec$, much longer than the \msec\
303: flares of RRATs.
304:
305: Compared to MSPs, this upper-limit to \fxfr\ is a factor 84 above the
306: observed ratio for (for example) the millisecond pulsar PSR~B1821-24
307: ($\fxfr = 5\tee{-14} \cgsflux \perval{\rm mJy}{-1}$;
308: \citealt{hoffman06}). Thus, this limit does not exclude an emission
309: mechanism for RRAT bursts with comparable X-ray to radio efficiency to
310: MSPs emission mechanism.
311:
312: The claim of measured $P-\dot{P}$ values (M06) which imply magnetic
313: field strengths approaching those of magnetars, makes comparison
314: between the RRAT-bursts and those from magnetars a natural one, with
315: the expectation that the emission mechanisms may be similar. The
316: absence of X-ray/radio burst flux limits for the \msec\ X-ray bursts
317: of magnetars
318: \citep{gogus99,gogus01,gavriil02,kaspi03,gavriil04,woods05} precludes
319: direct comparison with the emission mechanism powering those sources.
320: However, a radio observation of such would be valuable to compare with
321: the derived \fxfr\ limits for RRATs. Taking 1E~2259+586 as an example
322: \citep{gavriil04}, with median duration of $\sim$100 \msec, and peak
323: fluences of \ee{-9} \erg\perval{\rm\,cm}{-2}, magnetar bursts would
324: produce coincident radio bursts of $\geq2500\mJy$, if the magnetar
325: bursts were due to a similar emission mechanism as RRATs. This radio
326: flux is comparable to the brightest RRAT radio bursts yet observed
327: (M06), and are therefore readily detectable, yet no such bursts have
328: been reported. However, the burst rate of the \msec\ X-ray bursts of
329: magnetars is not constant in time; whether this is true for the radio
330: bursts of RRATs has not be described in the reported observations
331: (M06). On the other hand, a non-detection of radio bursts from
332: magnetars to flux limits below this value would require a different
333: \fxfr\ ratio for the \msec\ bursts from magnetars, and may therefore
334: imply a distinct emission mechanism.
335:
336: A burst flux limit $F_{\rm lim}<$1.5\tee{-8} \cgsflux (2-10 keV),
337: averaged over 3\,\msec, combined with an average burst rate of 228/13
338: hr (M06), results in time-average burst flux of
339: $<$2.2\tee{-13} \cgsflux\ (2-10 keV), or, extrapolating to the energy
340: band for the X-ray counterpart \citep{reynolds06} $<$2.1\tee{-13}
341: \cgsflux (0.5-8 keV). The time-average burst X-ray flux is a factor
342: of 10 less than the persistent X-ray counterpart in this pass-band
343: \cite{reynolds06}. Thus, X-ray bursts related to the radio
344: bursts of this RRAT are energetically unimportant compared to the
345: persistent X-ray flux. In addition, the time-average burst X-ray
346: efficiency $\eta=L_{X}(0.5-8\,\keV)/\dot{E}$, compared with the
347: reported dipolar emission $\dot{E}=24.94(5)\tee{31} \cgslum$
348: (M06) is $\eta\leq7\tee{-3} (d/3.6\,kpc)^2$.
349:
350: It is unlikely that great improvement will be made in the X-ray flux
351: limit presented here using existing instrumentation. An improved flux
352: limit cannot be obtained with {\em Chandra}: 2 counts in the HRC-S
353: detected within 2\,\msec\ corresponds to 3\tee{-8} \cgsflux
354: (extrapolated to 2-10 keV). On-source \rxte/PCA observations could
355: improve on the present flux sensitivity by a factor of $\times2$.
356:
357: \acknowledgements
358:
359: RER acknowledges support from NSERC through the Discovery Grant
360: program. RER acknowledges useful conversations with Marjorie
361: Gonzalez, Margaret Livingstone, Vicky Kaspi and Andrew Cumming.
362:
363: %\bibliographystyle{astro} \bibliography{complete}
364:
365: \begin{thebibliography}{}
366:
367: \bibitem[\protect\astroncite{{Blackburn}}{1995}]{ftools}
368: {Blackburn}, J.~K., 1995,
369: \newblock in R.~A. {Shaw}, H.~E. {Payne}, \& J.~J.~E. {Hayes} (eds.), {\em ASP
370: Conf. Ser. 77: Astronomical Data Analysis Software and Systems IV}, p. 367
371:
372: \bibitem[\protect\astroncite{{Gavriil} {\rm et~al.\/}}{2002}]{gavriil02}
373: {Gavriil}, F.~P., {Kaspi}, V.~M., \& {Woods}, P.~M., 2002,
374: \newblock {\em \nat} { 419}, 142
375:
376: \bibitem[\protect\astroncite{{Gavriil} {\rm et~al.\/}}{2004}]{gavriil04}
377: {Gavriil}, F.~P., {Kaspi}, V.~M., \& {Woods}, P.~M., 2004,
378: \newblock {\em \apj} { 607}, 959
379:
380: \bibitem[\protect\astroncite{{G{\"o}{\u g}{\"u}{\c s} } {\rm
381: et~al.\/}}{1999}]{gogus99}
382: {G{\"o}{\u g}{\"u}{\c s} }, E., {Woods}, P.~M., {Kouveliotou}, C., {van
383: Paradijs}, J., {Briggs}, M.~S., {Duncan}, R.~C., \& {Thompson}, C., 1999,
384: \newblock {\em \apjl} { 526}, L93
385:
386: \bibitem[\protect\astroncite{{G{\"o}{\u g}{\"u}{\c s}} {\rm
387: et~al.\/}}{2001}]{gogus01}
388: {G{\"o}{\u g}{\"u}{\c s}}, E., {Kouveliotou}, C., {Woods}, P.~M., {Thompson},
389: C., {Duncan}, R.~C., \& {Briggs}, M.~S., 2001,
390: \newblock {\em \apj} { 558}, 228
391:
392: \bibitem[\protect\astroncite{{G{\"u}del}}{2002}]{guedel02}
393: {G{\"u}del}, M., 2002,
394: \newblock {\em \araa} { 40}, 217
395:
396: \bibitem[\protect\astroncite{{Guedel} \& {Benz}}{1993}]{guedel93}
397: {Guedel}, M. \& {Benz}, A.~O., 1993,
398: \newblock {\em \apjl} { 405}, L63
399:
400: \bibitem[\protect\astroncite{{Hoffman} {\rm et~al.\/}}{2006}]{hoffman06}
401: {Hoffman}, K., {Rutledge}, R.~E., {Fox}, D.~B., {Gal-Yam}, A., \& {Cenko},
402: S.~B., 2006,
403: \newblock {\em \apj}, submitted, astro-ph/0609092
404:
405: \bibitem[\protect\astroncite{{Jahoda} {\rm et~al.\/}}{2006}]{pca2}
406: {Jahoda}, K., {Markwardt}, C.~B., {Radeva}, Y., {Rots}, A.~H., {Stark}, M.~J.,
407: {Swank}, J.~H., {Strohmayer}, T.~E., \& {Zhang}, W., 2006,
408: \newblock {\em \apjs} { 163}, 401
409:
410: \bibitem[\protect\astroncite{{Kaspi} {\rm et~al.\/}}{2003}]{kaspi03}
411: {Kaspi}, V.~M., {Gavriil}, F.~P., {Woods}, P.~M., {Jensen}, J.~B., {Roberts},
412: M.~S.~E., \& {Chakrabarty}, D., 2003,
413: \newblock {\em \apjl} { 588}, L93
414:
415: \bibitem[\protect\astroncite{{McLaughlin} {\rm et~al.\/}}{2006}]{mclaughlin06}
416: {McLaughlin}, M.~A., {Lyne}, A.~G., {Lorimer}, D.~R., {Kramer}, M., {Faulkner},
417: A.~J., {Manchester}, R.~N., {Cordes}, J.~M., {Camilo}, F., {Possenti}, A.,
418: {Stairs}, I.~H., {Hobbs}, G., {D'Amico}, N., {Burgay}, M., \& {O'Brien},
419: J.~T., 2006,
420: \newblock {\em \nat} { 439}, 817
421:
422: \bibitem[\protect\astroncite{{Reynolds} {\rm et~al.\/}}{2006}]{reynolds06}
423: {Reynolds}, S.~P., {Borkowski}, K.~J., {Gaensler}, B.~M., {Rea}, N.,
424: {McLaughlin}, M., {Possenti}, A., {Israel}, G., {Burgay}, M., {Camilo}, F.,
425: {Chatterjee}, S., {Kramer}, M., {Lyne}, A., \& {Stairs}, I., 2006,
426: \newblock {\em \apjl} { 639}, L71
427:
428: \bibitem[\protect\astroncite{{Swank} {\rm et~al.\/}}{1996}]{pca}
429: {Swank}, J.~H., {Jahoda}, K., {Zhang}, W., \& {Giles}, A.~B., 1996,
430: \newblock in M.~A. Alpar, U. Kiziloglu, \& J. Van~Paradijs (eds.), {\em The
431: Lives of the Neutron Stars}, NATO ASI Ser C., 450), Boston: Kluwer
432:
433: \bibitem[\protect\astroncite{{Woods} {\rm et~al.\/}}{2005}]{woods05}
434: {Woods}, P.~M., {Kouveliotou}, C., {Gavriil}, F.~P., {Kaspi}, V.~M., {Roberts},
435: M.~S.~E., {Ibrahim}, A., {Markwardt}, C.~B., {Swank}, J.~H., \& {Finger},
436: M.~H., 2005,
437: \newblock {\em \apj} { 629}, 985
438:
439: \end{thebibliography}
440:
441:
442: \clearpage
443:
444:
445: % Fig 0
446: \begin{figure}[htb]
447: \caption{ \label{fig:limits}
448: Upper-limits on burst rate as a function of flux, averaged on a 3\,\msec\
449: (solid line) and 0.3\,\msec\ timescales, from \rrat\ (see
450: \S~\ref{sec:results} for discussion).
451: }
452: \end{figure}
453:
454:
455:
456: % Figure 0
457: \clearpage
458: \pagestyle{empty}
459: \begin{figure}[htb]
460: \PSbox{f1.eps hoffset=-80 voffset=-80}{14.7cm}{21.5cm}
461: \FigNum{\ref{fig:limits}}
462: \end{figure}
463:
464: \clearpage
465:
466:
467: \clearpage
468:
469:
470:
471: \begin{deluxetable}{lccrcr}
472: \tablecaption{\rxte\ Observations of \rrat\ \label{tab:obs} }
473: \tablehead{
474: \colhead{Start Time (TT)} &
475: \colhead{End Time (TT)} &
476: \colhead{} &
477: \colhead{} &
478: \colhead{} &
479: \colhead{ $\langle I \rangle$ (c/s)$^a$} \\
480: \colhead{(dd/mm/yy hh:mm)} &
481: \colhead{(dd/mm/yy hh:mm) } &
482: \colhead{(ksec)} &
483: \colhead{\# PCUs$^b$} &
484: \colhead{$N_{\rm phot}$} &
485: \colhead{(1.5-10 keV)}\\
486: }
487: \startdata
488: 06/06/98 12:50 & 06/06/98 13:42 & 3.16 & 4 & 155048 & 49.07 \\
489: 06/06/98 14:25 & 06/06/98 15:25 & 3.63 & 4 & 183683 & 50.57 \\
490: 06/06/98 16:02 & 06/06/98 17:01 & 3.53 & 4 & 182342 & 51.57 \\
491: 06/06/98 17:42 & 06/06/98 18:37 & 3.31 & 4 & 159157 & 48.05 \\
492: 08/06/98 09:39 & 08/06/98 10:25 & 2.73 & 4 & 124533 & 45.52 \\
493: 08/06/98 11:15 & 08/06/98 12:01 & 2.73 & 4 & 119602 & 43.71 \\
494: 08/06/98 12:51 & 08/06/98 13:43 & 3.13 & 4 & 143406 & 45.73 \\
495: 08/06/98 14:27 & 08/06/98 15:26 & 3.55 & 4 & 167412 & 47.13 \\
496: 08/06/98 16:03 & 08/06/98 17:03 & 3.61 & 4 & 172435 & 47.69 \\
497: 08/06/98 17:43 & 08/06/98 18:40 & 3.39 & 4 & 161687 & 47.67 \\
498: 10/06/98 12:54 & 10/06/98 13:45 & 3.02 & 4 & 129359 & 42.78 \\
499: 10/06/98 14:28 & 10/06/98 15:27 & 3.56 & 4 & 158155 & 44.33 \\
500: 10/06/98 16:04 & 10/06/98 17:05 & 3.64 & 4 & 158619 & 43.48 \\
501: 10/06/98 19:21 & 10/06/98 20:17 & 3.34 & 2 & 140593 & 42.06 \\
502: 10/06/98 20:58 & 10/06/98 21:53 & 3.26 & 4 & 131917 & 40.42 \\
503: 10/06/98 22:34 & 10/06/98 23:14 & 2.43 & 4 & 97358 & 40.03 \\
504: 12/06/98 11:18 & 12/06/98 12:02 & 2.67 & 3 & 98495 & 36.88 \\
505: 12/06/98 12:53 & 12/06/98 13:46 & 3.21 & 4 & 134874 & 41.94 \\
506: 12/06/98 14:29 & 12/06/98 15:28 & 3.55 & 4 & 120253 & 33.86 \\
507: 12/06/98 16:05 & 12/06/98 17:05 & 3.61 & 3 & 113601 & 31.42 \\
508: 12/06/98 17:41 & 12/06/98 18:41 & 3.63 & 3 & 111473 & 30.69 \\
509: 12/06/98 19:19 & 12/06/98 20:17 & 3.50 & 3 & 105962 & 30.24 \\
510: \enddata
511: \tablenotetext{a}{Average countrate during observation for PHA
512: channels $\leq$27.}
513: \tablenotetext{b}{Number of active PCUs at observation start time.
514: PCUs turn on and off during the course of the observations.}
515: \end{deluxetable}
516:
517:
518:
519:
520: \end{document}
521:
522: To develop msec burst discovery methods, and remove instrumental
523: effects, we make use of Observation 70801; this makes use of 6
524: different pointing directions on the sky for background purposes
525: (called BACKGROUND\#N, N=[1,6]), and includes slew data (we do not
526: exclude data based on GTI). We exclude observations with $<$3 PCUs
527: active at the time of observation start. There are 876ksec of data in
528: 992 separate observations (including slews) with integrations between
529: 31 and 4300 ksec. We examine {\tt GoodXenon\_2s} data.
530:
531: \subsection{Analysis Verification}
532:
533: We performed this analysis on source-less background fields -- 275 hrs
534: of data in Observation ID 70801 during which we found 97 HV
535: breakdown events, and no triggers.
536:
537: