UvA-DARE (Digital Academic Repository)

Photometric observations of faint cataclysmic variables

Howell, S.B.

Publication date 1995

Link to publication

Citation for published version (APA): Howell, S. B. (1995). Photometric observations of faint cataclysmic variables.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:25 Sep 2021 Chapterr 6

CCDD Time Resolved Photometry of Faintt Cataclysmic Variables IV

S.. B. Howell, P. Szkody. T. Kreidl. and D. Dobrzycka Publ.Publ. of the Ast. Soc. of the Pac. 1991, 103, 300

ABSTRACT T

Wee observed fifteen faint stars, known or suspected to be cataclysmic variables, obtaining ('CDD time resolved photometry in V', B, and the near-1R. The stars are: \VX Cet, CP Eri. XZZ Eri. V592 Her, TT Ind, AO Oct, BE Oct. CT Ser. V227G Sgr. KK lei. DR CMa. R\V EMi,, PG0917+342, PG2240+19:Ï, and 2138 - 15:5. All of the stars were observed for - 2 or moree hours, with the average time series length being over 3 hours. We find orbital periods forr the stars KK Tel and 2138 — -1-r>;3 and a likely period confirmation for RW EMi. ('I' Eri iss a probable DQ Her star, showing an optical modulation of 29 minutes. Nine other stars showw photometric evidence which likely confirms them as cataclysmic variables, while Un- truee natures of PG0917+342 and PG224Ü+193 are not yet revealed.

6.11 Introduction

Thiss paper is the fourth in a series of our continuing project to obtain ('CD time resolved photometryy of faint cataclysmic variables (CVs) (Paper I: Howell and Szkody 19N8; Paper II:: Szkody et at. 1989: Paper III: Howell et al. 1990). Cataclysmic variables (CVs) are

99 9 100 0 ('CD('CD Time-Resolved Photometry of Faint CVs closee binaries in which a mass-losing secondary (a late main-sequence dwarf) is in orbit with aa white-dwarf primary. Typical dimensions for these systems are of the order of a solar diameter,, and accordingly they have periods that are very short; a little over 1 hr to about. 155 hours. The material that is lost from the secondary usually accumulates in a disk around 11 he primary with eventual accretion onto its surface. This transfer of material to the disk can bee responsible for flickering (random low-level variations of order 0.1-1.0 mag), and periodic variationss due to viewing the hot spot created at the stream/disk interface. An increase in thee transfer rate of material from the secondary to the disk, or a disk instability leading too increased transfer of material through the disk, is thought to be the cause of the usual 2-55 mag outbursts observed in the dwarf novae type of CVs (see the review by Wade and Wardd 1985). All dwarf novae with periods below the period gap show, at times, enhanced outburstss called superouthursts. During the superoutbursts. there are photometric humps presentt at a period a few percent longer than the orbital period. These stars belong to a subgroupp of CVs known as the SI Crsae Majoris stars (reviewed by Warner 1985).

Thee stars we observe in this continuing project generally have little or no previous informationn other than some hint of being likely candidates for CVs. The goals of obtaining ('CDD photometry on such binaries and determining their colors are 1) to search for variability andd periodicities related to the orbital periods and the spin periods of magnetic white dwarfs; andd 2) to look for long- term changes in the quiescent properties of these stars (see Paper 111). Thee major focus of this study is on stars with evidence for short periods (less than 4 hours), andd CVs that an1 at high galactic latitudes as described in Howell and Szkody (1990). Other faintt unstudied and/or interesting systems are also included. Detailed analysis of interesting systemss and synthesis of our collected data are published separately.

('singg ('CDs as our detectors, we can gather time series data with one to ten-minute timee resolutions for stars from ~ 14 magnitude down to magnitude 20-22 with 1-2 meter classs telescopes. For the faintest stars (V' > 17 — 18) we are usually providing the lirst such dataa available. Papers 1 through IV contain data on a total of 52 stars and over 100 time seriess dat a sets.

6.22 Observations

Thee data presented here were obtained with the Perkins 1.8-m telescope of the Ohio Statee and Ohio Wesleyan Universities at the Lowell Observatory, the Kitt Peak National Observatoryy No. 1 0.9-m telescope, and the Cerro Tololo Inter-American Observatory 0.9- mm telescope. The data from Kitt Peak and CTIO were reduced either with the "cdphot" packagee available on the instrument computer at the telescope or with the APPHOT package 6.3.6.3. RESULTS 101 1

inn 1RAF '. The Lowell data were reduced with the IRAF/APPfIOT package. Table (U gives ann observing log for the data presented here.

Ourr general observing policy for these types of data is to observe a few photometric standardss each night to determine rough offsets between instrumental and real magnitudes. Wee obtain single color measurements directly before or after our time series data. The standardss are then used to give the average magnitude values listed in Table (i.1. with an errorr of ~ 1 mag. The time series data sets were reduced and have their errors determined ass given in Howell e.t al. (1988).

Alll the data sets were searched for periodic modulations using the technique of Phase Dispersionn Minimization (yield) developed by Stellingwerf (1978). The PDM technique pro- videss the user with a test statistic for eacli determined period that yields a confidence level viaa an (two-sided) F-test, We indicate that, no periods were seen if our PDM analysis fails too reveal any period at all with a confidence level of > 70 - 80%. Between ~ 80% and 90% confidencee level (CL), we generally will mention the result, as it may possibly be of some interestt or get confirmed later. We define a period to be real if it gives an acceptable (usually >> 9')%.) confidence level, and we believe it to be stable if it is seen on more than one night at thiss high OL. Table 6.2 gives a list of the periods seen in our l-r) stars from this paper which falll into the L'real and stable'' category. Many CVs do, however, show nightly changes in averagee magnitude level and flickering characteristics, which may lead to erroneous "periods" beingg determined. The less confident periods, that may or may not be real, are listed in the individuall discussion for each star presented in the next section.

6,33 Results

6.3.11 WX Ceti

Paperr II presented our initial work on this star. A possible period of 5 minutes (inn the near-IR) was seen, but with low confidence since the data set length was short. We presentt here two new data sets taken in V and B (see Figure (i.1). Our V' data set shows no indicationn of any periods and shows simply a flickering light curve. Our B data set. however, showss indications of periods at 8 and 7 min. These periods are at confidence

[RAFF is distributed by National Optical Astronomy Observatories, which is operated by the Association off 1'niversitios for Research in Astronomy. Inc., under cooperative agreement with the National Science Foundation n 102 2 CCDCCD Time-Resolved Photometry of Faint CVs

Tab lbee 6.1: Observing gLo g g Star r ITT Dato ITT Start" N'1 1 C Cdt 6 6 TTh h CCDr r Filter" " Mag' ' ^ ^ \VXX (Vi OlSepSO O 08:29:13 3 60 0 180 0 23 3 3.4 4 TI6 6 B B 17.8 8 0.02 2 08:28:40 0 V V 17.6 6 03Xov89 9 06:16:40 0 29 9 300 0 13 3 2.5 5 RCA A V V 17.7 7 0.03 3 (TT Eri 04Sep89 9 09:16:37 7 30 0 320 0 23 3 2.9 9 TI6 6 B B 19.3 3 0.05 5 09:12:30 0 V V 19.7 7 26Aug90 0 08:25:56 6 21 1 300 0 29 9 1.9 9 TI3 3 V V 19.8 8 0.03 3 XZZ Eri 04Xov89 9 08:14:20 0 55 5 220 0 13 3 3.8 8 RCA A V V 18.7 7 0.06 6 V5922 Her 20Apr90 0 09:06:10 0 15 5 400 0 13 3 1.7 7 RCA A \VR R 0.10 0 09:05:20 0 V V 20.5 5 Trrr Ind 277 Aug!)» 00:50:14 4 100 0 120 0 29 9 4.1 1 TI3 3 V V 0.12 2 23:52:00 0 B B 16.8 8 23:58:00 0 V V 16.6 6 28Aug90 0 00:01:00 0 R R 16.1 1 AC)) Oct 25Aug90 0 04:10:11 1 10 0 360 0 29 9 1.1 1 T13 3 V V 21.1 1 0.14 4 03:42:18 8 B B 21.0 0 03:55:00 0 V V 20.9 9 04:02:00 0 R R 20.7 7 23:44:33 3 R R 20.2 2 23:51:38 8 V V 20.9 9 23:58:34 4 B B 20.6 6 26Aug90 0 00:06:37 7 37 7 360 0 29 9 4.0 0 TI3 3 V V 20.9 9 0.09 9 BEE Oct. 27Aug90 0 05:48:26 6 45 5 300 0 29 9 4.1 1 TI3 3 V V 19.4 4 0.03 3 09:47:05 5 B B 19.6 6 09:53:19 9 R R 19.3 3 (TT Ser 14Apr90 0 08:59:10 0 25 5 180 0 13 3 1.3 3 RCA A V V 16.4 4 0.06 6 08:43:41 1 B B 16.2 2 Y22766 Sgr 29Aug90 0 02:53:05 5 28 8 300 0 29 9 2.6 6 TI3 3 V V 20.1 1 0.07 7 02:40:41 1 B B 19.9 9 02:46:45 5 R R 20.1 1 KKK To] 28Aug90 0 00:26:50 0 B B 19.5 5 00:21:11 1 86 6 180 0 29 9 5.7 7 TI3 3 X' X' 19.2 2 0.03 3 00:08:37 7 B B 19.4 4 00:14:52 2 R R 18.8 8 29Aug90 0 23:46:01 1 41 1 180 0 29 9 2.4 4 TI3 3 B B 19,4 4 0.05 5 23:41:14 4 V V 19.3 3 6.3.6.3. RESULTS 103 3

Tablee 6.1: Observing Log 6 6 r r Star r 1'TT Date ITT StarC X C C dC C T* * CCD Filter'' ' Mag--

''' B, V. and R filters are Johnson filters; WR (Wide R) has Xc = 700!)Aand FVVHM == 2601 A. ee Standard stars were observed nightly, usually near the beginning or end of a time series.. Xo corrections for color terms or airmass were made. Errors in absolute magnitudess are ~ 0.1 mag (see Howell et al. 1988). The listed magnitude for the timee series is the average over the time series. The other magnitudes are single measurementss made directly before or after the time series. WR magnitudes are estimatedd from spectrophotometry standards. // a calculated from Eq. 4, Howell et al. 1988. 99 the CCD frames were co-added in this time series to yield the equivalent of 500 secondd exposures. 104 4 CCDCCD Time-Resolved Photometry of Faint CVs

.. 2 I—i—r -i—i—i—i—r—r--

--

3 3 > > n n << .6

WXX Cet lSep89 UT II L_J 1 I I I I I 1 1 1 100 11 TIMEE (HRS)

Fig.. 6.1.- The light curve of WX Cet taken in (a) B on 1 Sep 1989 and (b) V on 3 Nov 1989.. Note the amplitude is smaller in V and the light curves appear very dissimilar. Each pointt represents a differential magnitude measurement relative to a comparison star from the samee CCD frame. The points are plotted at the mid-exposure time of each integration with !! he first point being at the time given in Table 6.1 for the UT start. The magnitude listed in Tablee 6.1 for each series corresponds to the average magnitude seen during the observations. Thee error bars shown are r and are displayed with a sample data point. They are equal too the sigma value listed in Table 6.1. If no error bar is shown, the error is smaller than the sizee of the plotted symbol. 6.3.6.3. RESULTS 105 5 levelss of 80 and 85%, respectively. These periods are certainly shorter than that of 6.98 hr suggestedd by Van Paradijs et ai (1990), based on spectroscopic studies.

Recently,, O'Donoghue (1990) observed \VX Cet during superoutburst. His determined superhumpp period is likely to be 73, 76. or 81 min, leading to a probable orbital period (see Warnerr 1985) of 72-80 minutes. The periods seen in our B data are in general agreement with thee orbital period found by O'Donoghue. with the 44-minute period being a likely harmonic. Thee superhunips observed and our photometric evidence almost certainly indicate a short orbitall period for WX Cet, likely to be near 5 min.

Wee note that there may be a decrease in photometric flickering amplitude with color (0.33 mag in B and 0.1 mag in R [see Paper II]), although our data were obtained more than aa year apart. In order to determine the source of this photometric modulation, multicolor photometryy would be useful. These observations could differentiate between a hot spot structuree (decreasing amplitude in the red) and general flickering.

6.3.22 CP Eridani

Inn paper II, we presented a time series light curve for this star. It was seen to be at an averagee magnitude of V = 17.8 and showed a slight increase of 0.2 mag over the 2.7 hours off monitoring. There was no indication of any periodic modulations. We observed CP Eri againn in September, 1989 and in August, 1990. During these observations, CP Eri was at a magnitudee of 19.7 and showed a very distinct modulation in both B and V light (Figure 6.2). Thee B data show large amplitude (0.2-0.4 mag) modulations and a clearly defined period off 3 min, while the V data had smaller (0.1-0.2 mag), less well defined modulations yieldingg a period of 5 min. The V data set also shows a strong harmonic at 13.6Ü.5 min. .

Tablee 6.2: Periods Evident in Data j 3 3 Object t P( rio( d d(min ) ) S Sgma (min) ) C.L.' ' (%) ) Periodd Type CPP Eri 29 9 1.4 4 99 9 magnetic c KKK Tel 121 1 5 5 95 5 orbital l RWW ('Mi 113 3 10 0 95 5 orbital l h h 21388 -453 93 3 11 1 mm orbital l aa Confidence Level, as given in Stellingwerf (1978). 66 See text. 106 6 CCDCCD Time-Resolved Photometry of Faint CVs

100 11 TIMEE (UT HOURS)

11 1 11 1 1 1 I 1 11 1 1 1 I 1 1 1 I 1 TT r T" CPP Eri 26Aug900 UT H H 55 4.1 // j > > 1 V1 i|//V V

-- 1 1 V33 V9.5 TIMEE CHRS)

Fig.. 6.2.— Lightrurves in (a) B and (b)V' for the DQ Her system CP Eri. Sec Figure 6.1 captionn for a complete description. 6.3.6.3. RESULTS 107 7

11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ii i | i i

-- . m+ -- -- - * i i

? ? --

XZZ Eri 4Nov89 UT >> i i i 1 i i i i 1 i i i i 1 i ,, , I . . 100 11 TIMEE (HRS)

Fig.. 6.3.— Lightcurve of XZ Eri taken on 4 Nov 1989. See Figure 6.1 caption for a complete description. .

III I -i—i—i—i—i—r r V5922 Her 20Apr90 UT

5 5

-JJ [—l—I 1—I—I I I I I I 1 I 1 I I 1 !_ 3.55 10 10.5 TIMEE (HRS)

Fig.. 6.4.— Lightcurve for V592 Her taken 20 Apr 1990. See Figure 6.1 caption for a complete description. . 108 8 CCDCCD Time-Resolved Photometry of Faint CVs

ii i I i i i i I i i i i I i i i i I ' ' ' i I i TTT Ind 27Aaig90 UT

m

'' 1111 i i—11 122 3 TIMEE CHRS)

Fig.. 6.5.-- Lightcurve of TT Ind taken 27 Aug 1990. See Figure 6.1 caption for a complete description. .

f T 1 T [ ! ! 1 I | i i i i | I i ii i 1 i i

5 5 -- >> 1. s mm m : : m m i

AOO Oct 26Aug90 UT ii i i i 1 i i i i 1 i i i i 1 i i >> > I i i 122 3 TIMEE CHRS)

Fig.. 6.6.— Lightcurve of AO Oct taken on 4 Nov 1989. See Figure 6.1 caption for a complete description. . 6.3.6.3. RESULTS 109 9

_ _ I1 1 11 1 | 1 1 1 1 f—1—T 1—1 [—1 ii i i |_ .. 4 ** "

"

R R X X ff " wTwT _ _ mm m ~ ' ' " " mm — " " mm ~ ~ 1 1 -- "-- - rr -

1.2 2 BE EOc tt 27Aug90 UT | | ii 1 i i 11 1 1 1 1 1 1 1 1 1 1 1 1 ,, , , i" 77 8 9 TIMEE CHRS)

Fig.. 6.7.— Lightcurve of BE Oct taken on 4 Nov 1989. See Figure 6.1 caption for a complete description. .

^^—I—i—i—i—i—I—i—i—i—rII ' ' ' ' I —I—i—i—i—i—I—i—i—i—r

> >

V22766 Sgr 29Aug90 UT J—ii I i i i i L_i i i i I i ' i i I i 33 3. S + 4. S S TIMEE (HRS)

Fig.. 6.8.— Lightcurve of V2276 Sgr taken on 4 Nov 1989. See Figure 6.1 caption for a completee description. 110 0 CCDCCD Time-Resolved Photometry of Faint CVs

V VV ,

KKK Tel 28Aug90 UT —ii 1 1 1 1 1 1 i 22 4 TIMEE (HRS) — —

2 2

u u SS 2. 2 •• + VT T

en n

KKK Tel 29Aug90 UT -JJ I L. 2 2 TIMEE (HRS)

Fig.. 6.9. - - V and B light curves for the star KK Tel. These data were taken on consecutive nightss and generally have the same appearance. See Figure 6.1 caption for a complete description. . 6.3.6.3. RESULTS 111 1

Wee conclude that the Paper II observation was probably made during a rise to outburst duringg which the light curve modulations presented here were completely masked. CP Eri's determinedd period of 29 min, being too short to be an orbital period (Nelson et al. 1985), suggestss that this star is likely to be a new DQ Her system.

6.3.33 XZ Eridani

Noticedd by Shapley and Hughes (1934) to be a variable with a magnitude range of from 14.66 to fainter than 17.5, this star was a likely candidate to be a CV. We obtained ~ 4 hr of timee series photometric data, during which time it was at a mean magnitude of V — 18.7. Thee light curve (Figure 6.3) shows flickering at the 0.4 mag level, characteristic of a (-V, butt no indications of periodic modulations were found. A radial-velocity study of this star att minimum is likely to be needed to determine its orbital period.

6.3.44 V592 Herculis

Thiss star is listed as a nova of type NA in the GCVS and as type UG? or XND? in Duerbeckk (1987). It was discovered by Richter (1968) on Sonneberg plates and found to havee a magnitude range of mpg = 12.3 to greater than 20. The light curve during decline appearedd similar to that of the X-ray nova V616 Monocerotis, and the object is blue in color att maximum. Duerbeck suggests that these latter two properties make classification as a classicall nova unlikely.

Wee observed the entire field shown in Duerbeck, including his marked candidate stars 11 and 2, and found that only star 1 showed any significant variability during the 1.7 hour timee series. Figure 6.4 shows the light curve obtained when the star (Duerbeck's candidate 1)) was at a V magnitude of 20.5. No periodic modulations were seen, but large amplitude flickeringg of 0.4-0.6 mag is clearly visible. We therefore feel fairly certain that candidate 1 iss the star associated with V592 Her and the "nova'' observed in 1968. Further study at minimumm is needed to determine the true nature of this star.

6.3.55 TT Indi

Thiss star was discovered at magnitude 14.0 by Gessner and Meinunger (1974). We observedd this star when at minimum, V = 17.0. This observation gives the first measurement off this star at quiescence and shows it to have an outburst amplitude of at least 3 mag. 112 2 CCDCCD Time-Resolved Photometry of Faint CVs

Ourr 4 hr time series light curve (Figure 6.5) taken in August, 1990, showed typical ('V behaviorr with flickering of ~ 0.2 mag, but no indication of any periodic modulations.

6.3.66 AO Octantis

Listedd in the GCVS as a UG: type variable, this ('V was discovered by Gessner and

Meinungerr (1974) during an outburst at magnitude 13.5. Their quiescent value listed mpg = 21,, which agrees well with our determined value of V — 20.9. The outburst amplitude of thiss star is quite large (7.5 mag), making it a very interesting system to study.

Twoo photometric data sets of 1 and 4 hr (see Figure 6.6) were obtained. Large amplitude flickeringg of 0.5-0.65 mag was observed but no modulations of a periodic nature were found.

6.3.77 BE Octantis

BEE Oct is listed in the GCVS as a UGSC type variable with a magnitude range of 15.4 too > 17.5. Petit (1970) noted an apparent extended outburst time and suggested that this objectt is a dwarf nova of the SU UMa subclass.

Ourr 4 hr light, curve (Figure 6.7), obtained when BE Oct was at a V' magnitude of 19.4. showedd 0.2 mag flickering, typical for CVs in quiescence, but presents no orbital or other periodicc modulations.

6.3.88 V2276 Sagittarii

Petitt (1970) describes this star as a l.'G type object with an outburst period of < 30 days.. The GCVS lists it as having a maximum magnitude of 14.4. with minimum being fainterr than 16.7. Our observations, made at quiescence, showed it to have a V magnitude off 20.1. A single time series data set (Figure 6.8) reveals typical flickering usually associated withh a OV at minimum, but shows no periodic modulations.

6.3.99 CT Serpentis

Inn Paper III. we indicated that CT Ser might show periodic modulations of order 13 andd 65 minutes. We therefore re-observed it to see if these possible periods were still present, therebyy indicating their reality. 6.3.6.3. RESULTS 113 3

Ourr new photometric data set shows a light curve that is essentially flat, with only small amplitudee (< 0.1 mag) modulations. A period search found no indication of any period, particularlyy the ones suggested in Paper III. We therefore conclude that either the periods seenn were quasi-periodic in nature or were simply flickering modulations masquerading as periods. .

6.3.100 KK Telescopii

Notedd as a variable blue object by Hoffmeister (1963), KK Tel is listed as having a magnitudee range from mpg = 13.5 to ~ 19.7. Hoffmeister suggested that the nature of the variationss he observed indicated that this star is a likely member of the UG class of variables.

Ourr two photometric light curves (Figure 6.9) show quite clear evidence for a periodic modulation.. The two data sets shown were taken on consecutive nights in V and B, respec- tively.. The data from the first night show a period of 121 5 min. The two nights together gavee the same period with a CL of 95%, but with a larger error 4 min). The different shapee of the hump seen in B is the likely cause of this larger error. The 121 min period extrapolatedd to the second night predicts the peak midpoint to be at 0.82 hr (see Figure 6.9b),, while the real midpoint appears at about 1.06, the difference being 14.4 minutes. This modulationn is very likely to be from a hot spot, thereby indicating the orbital period of KK Tel,, with the difference in phasing on the two nights possibly being due to hot spot changes inn size or location. We see that even within a single night (see Figure 6.9a), the hot spot humpp shows different shapes and is not always symmetric.

Inn order to investigate the properties of our light curve data in detail, we have phased togetherr the V data obtained on the first night (Figure 6.10). These data may show some evidencee for an additional hot spot, offset by 0.5 in phase from the main (typical) hot spot. Additionall spots of this kind have been seen in a few other systems, such as DV UMa. Howell ftft al. (1988). Further extended photometric observations of KK Tel are planned in order to confirmm the phenomena discussed here.

6.3.111 DR Ursae Majoris

Romanoo (1979) discovered this star while searching near M101 for high galactic latitude variables.. He noted that it had a range of 17.5 to fainter than 18.2 and a B — V magnitude of +0.11 to +0.5. This evidence led him to the conclusion that this star is a likely CV candidate. 114 4 CCDCCD Ti me-Resolved Photometry of Faint f'V'.s

Csingg the chart provided by Romano, we looked for this object in June. 1989. and found noo object in the field at or near his marked position that was blue, or that showed significant, variationn down to a \' magnitude of 1S.5. Almost one year later, in April, 1990. we again lookedd for this star and found it at a magnitude of V' — 19.1. A re-examination of our originall data from approximately one year earlier revealed that the star was indeed present butt that, its magnitude was fainter, being V' = 20.2. Summing our original 1989 frames in groupss of two gave us poorer time resolution, but did allow us to construct a light curve withh reasonable signal to noise characteristics. Two of the light curves from 1989 and 1990 aree shown m Figure 6.11.

AA period search of each data set independently, as well as the two sets from April. 1990 combined,, revealed no significant periods. Our three time series data sets do. however, show >> 0.2 mag variations throughout. For DR. 1'Ma we presume a typical ('V nature can explain ourr observations.

6.3.122 RW Ursae Minoris

Paperr II showed the first time series light curve for this star. It indicated a r> minn periodicity in the near-IR, but confirmation was needed since we only had a single nights'' data. The modulation seen in the near-IR looked sinusoidal and, if it represented the changingg aspect of a Roche lobe filling secondary (ellipsoidal variations), it would indicate ann orbital period of ~2.'J4 minutes.

Wee present here new time series photometry, taken in B. in hopes of dist iijguishing be- tweenn the possible confusion of the true period being once1 or twice the observed photometric period.. Since this star has a very high declination f+ 77 degrees), it is difficult to study it for extendedd time periods due to the construction of most equatorial mounted telescopes. Our 2.11 hr B data set (see Figure (i.12). shows a modulation which has a similar peak-to-peak amplitudee and overall appearance to the near-IR data shown in Paper II. The B data show aa period of 0 min. consistent with the period of 117+5 min found in the data w»> presentedd in Paper II.

Iff the modulation seen is due to the ellipsoidal nature of the secondary, the correct explanationn of the photometric data indicates an orbital period for RW I'Mi of ~4 hr. However,, the non-sinusoidal shape and amplitude of the H data suggest its origin is nut the changingg aspect, of t he secondary, but a large bright portion of the disk (hot spot?), coming intoo view once each orbital period. Our best guess so far is that the orbital period of RW II Mi is ~1.9 hr. but we believe a better interpretation can result only from a simultaneous multicolorr or spectroscopic study at minimum. 6.3.6.3. RESULTS 115 5

6.3.133 PG0917+342

PG0917+3422 was found during the Palomar-Green Survey (PG Survey; Green et al. 1986)) and identified as a CV candidate by Green et al. (1982). It has also been suggested to bee the missing Nova Lynx 396 A.D. (Shara 1989). Green ü al. (1982) mention that the star- wass observed in both emission and absorption states and that while in an emission state, thee line strengths appear to look like those of a typical dwarf nova, although He lines are nott present.

Ourr single ~ 4 hr light curve (Figure 6.13) shows the star to be constant within the errorss throughout the entire data set. This type of light curve is atypical for dwarf nova in thatt it does not show the normal flickering behavior, but would be typical for an old nova, possiblyy indicating a thick disk. PG0917+342 does, however, show two dissimilar looking ~~ 0.1 magnitude humps in its light curve. These features are not greater than 3cr but their cohesivee appearance is compelling. If these humps represent an orbital modulation, then the orbitall period would be ~ 2 hr. Further photometric observations, particularly in B. are planned. .

6.3.144 PG2240+-193

Thiss star was also identified in the PG Survey and described by Green et al. (1982). The spectrumm Green et al. describe has Ho in emission but the higher Balmer lines in absorption. Theyy also mention that it might have a Fe II spectrum and that no He lines appear. We observedd this star for 3.5 hr and the light curve (Figure 6.14) is constant to within ~ 0.05 mag,, possibly indicating that PG2240+193 has a period longer than 3.5 hr.

6.3.155 2138-453

Hawkinss and Veron (1987) discovered this object during their search for faint variable quasars.. They present a spectrum (4000A-7000A) taken when 2138 - 453 was at B = 20.4. Itt looks fairly typical for a dwarf nova at minimum, with the Balmer lines and Hel A 5876 in emission.. The Balmer lines are double peaked, with Ho having a separation of 900 km sec-1. Wee therefore added this object to our observing list.

Ourr time series observations obtained in V and B are shown in Figure 6.15. The V data sett shows a fairly typical hot spot modulation and some marginal evidence for an additional humpp offset by 0.6 in phase. The system shows large flickering in both data sets. 116 6 ('CD('CD Time-Resolved Photometry of Faint CV.s

Eachh data set was searched for periods independently. We found a period of 1 min (CLL of 88%) in the V data set and 1 min (CL of 86%) in B. We then combined the data setss in hopes of better refining the period. What we found, however, was that each of the two periodss above remained and there1 was no indication of any single period common to buth sets.. Plots made of both nights' data phased together on either period are not convincing. butt show a much better fit to the periodic modulations found in \ data set. Also, taking thee 93-minute period and extrapolating it to the second night shows a reasonable fit to the twoo larger peaks seen. We therefore list the period of'2138 —453 as 1 I min in fable 6.3 andd believe it to he real, within our errors, based on its repeatability. I he large flickering seenn may mask a good determination of the true orbilal period, particularly on the second night.. Longer term multicolor light, curves are needed.

6.44 Summary

Wee have1 presented CCD photometry time series data for 11 new stars and further data forr 1 stars. Two new orbital periods have been found for the stars KK Tel and 2138 — 153. andd CP Eri is a likely DQ Her star. WX (Jet is likely to have an orbital period of 72 to 80 minutes,, and RW 1'M'fs orbital period is probably near 113 minutes.

Inn Paper III we pointed out the fact that some of our program CVs show temporal magnitudee changes at quiescence. DR 1'Ma is yet another example and gives us a hint of thee time scales involved as it changed by one magnitude on a time scale of no longer than 100 months. We also know that for the time period of 3 days (16 to 19 April 1990) it stayed withinn 0.1 mag of I' = 19.1.

Ourr four papers in this series so far have presented ('CD light curves for 52 previously unstudiedd si ars known or suspected to be cataclysmic variables. These data have consisted off 120 time series lijf lit curves. -17 of duration > 3 hr. Of the 52 different stars that have beenn observed, we have found 22 which show modulations indicative of orbital periods (10) orr magnetic pulses (3). Most of the other 30 stars have shown photometric evidence which supportss their inclusion as cataclysmic variables.

Wee are continuing our CCD photometry project of obtaining data on faint OYs and pur- suingg follow-up study on the interesting systems we have already discovered. Distances and radial-velocityy si udies are of high priority in order to ascertain t he true absolute magnitudes off the faint CYs and provide measurements of their intrinsic properties.

Wee would like to thank the staffs of the observatories we visited for their help, without References References 117 7 whichh this project could not be done. T. Kreidl acknowledges support from the Lowell Observatoryy Endowment. P. Szkody and S. Howell acknowledge support from NSF grant ASTT 89-15145. S. Howell also acknowledges support by a grant from NASA administered byy the American Astronomical Society. Planetary Science Institute is a non-profit division off Science Applications International Corporation. This is PSI Contribution No. 287.

References s

Duerbeck,, H. 1987. A Reference Catalogue and Atlas of Galactic Novae (Boston, Reidel)

Gessner,, H. k Meinunger, I. 1974, VSS, 6, Ho, 249

Green,, R.F., Ferguson, D., Liebert, J. k Schmidt, M. 1982,0 PASP, 94. 560

Green,, R.F., Schmidt, M., k Leibert, J. 1986, ApJ Suppl. 61, 305

Hawkins,, M.R.S. k Veron, P. 1987, AkA, 182. 271

Hoffmeister,, C. 1963, VSS, 6, No. 1

Howell,, S.B. k Szkody, P. 1990, ApJ, 356, 623

Howell,, S.B., Mason, K.O., Reichert, G.A., Warnock, A., k Kreidl, T. J. 1988, MNRAS, 233,, 79

Howell,, S.B. k Szkody, P. 1988, PASP 100, 224 (Paper I)

Howell,, S.B., Szkody, P., Kreidl, T.J., Mason, K.O., k Puchnarewicz, E.M. 1990, PASP 102, 7588 (Paper III)

Nelson,, L.A., Chan, W.Y., k Rosenblum, A. 1985, ApJ, 299,658

O'Donoghue,, D. 1990, IAU Circular #5006 and private communication

Petit.. M. 1970, CT, 86, 229

Richter.. G.A. 1968, IBVS, 293

Romano,, G. 1979, IBVS, 1674

Shapeley,, H. k Hughes, E.M. 1934. HA. 90, No. 4

Shara.. M. 1989, private communication

Stellingwerf,, R.F. 1978, ApJ. 224, 953

Szkody,, P., Howell, S.B., Mateo. M., k Kreidl. T.J. 1989. PASP 101, 899 (Paper II) 118 8 CCDCCD Time-Resolved Photometry of Faint ('Vs

Vann Paradijs. J.. Van der Klis. M., k Pedersen. H. 1990. AkA. 225. L6

Wade,, R.A., & Ward. M.J. 1985, Interacting Binary Stars, ed. .I.E. Pringle & H.A. Wade (Cambridge,, Cambridge University Press), p. 129

Warner,, B. 1985. in Inttracting Binarits, Ed. P. Eggleton «k- J.E. Pringle (Dordrerlit:Keidel). p.. 367 References References 119 9

-i—I—i—|—i—i—i—r r

2 2 11 2. E > >

22 B KKK Tel 28Aug90 UT II .... I _i_L L .5 1 1. s PHASEE (PER. 2 HRS)

Fig.. 6.10. - KK Tel V data from 28 Aug 1990 orbitally phased on a determined orbital periodd of 2 hours. Note the possible existence of a secondary hot spot hump, offset in phase fromm the primary by 0.45 in phase. See text for details. 120 0 CCDCCD Time-Resolved Photometry of Faint CVs

i—i—i—|—i—i—i—i—|—i—i—i—i—|| i i i i f DRR UMa 7JunB9 UT

•• • • • " • • > > << .6 "" -"'S

jj i i i i i i_ 33 10 11 1 TIMEE CHRS) II I I | I I I I | I I 11 ' ' I ' ' ' ' I

66 - *

•• • •• •

Sii i

DRR UMa 19Apr90 UT

66 7 B ' ' TIMEE (HRS)

Fig.. 6.11. - Lightcurves for DR UMa from (a) 7 Jun 1989 in V and (b) 19 Apr 1990 in R. Seee Figure 6.1 caption for a complete description. References References 121 1

—p p 11 I | l 1 > 1 1 I 1 1 1 [ 1 1 1-1 | 1 1 1 1 | l

1.. 2

1.. 1 .. _ -

i i

9 9 RWW UMi lSep89 UT 11 1 1 1 1 i. J 1 M 1 I 1 1 1 1 1 [ 1 1 1 1 1 l~ 5.. 5 S 6.. S 7 7. S TIMEE CUT HRS)

Fig.. 6.12.— B photometry or RW UMi taken on 1 Sep 1989. See Figure 6.1 caption for a completee description.

-i—I—i—I—i—r-- "|| i i i—i—|—i—i—i—r

pp -1 6 I I K K rtSflfli^" rtSflfli^" << -1. +

PG0917+3422 16Apr90 UT

ii i i i 1 JJ 1 1 1 I L. ++ 56 TIMEE (UTHRS)

Fig.. 6.13.— R light curve of PG0917+342 obtained on 16 April 1990. See Figure 6.1 caption forr a complete description. 122 2 CCDCCD Time-Resolved Photometry of Faint CVs

T—i—i—I—|—i—i—i—I—|—i—i—r-T—|—i—i—i—ii | r

*^Puu '****£&&$* > >

PG2240+1933 31Aug89 UT

_11 I I 1 1_ ii i i i 88 3 10 TIMEE CHRS)

Fig.. 6.14.— V light curve of PG2240+193 taken on 31 Aug 1989. On this particular night, thee observations were made during a period of passing thick clouds. The light curve has less dataa during these cloudy times, particularly near 8.3 and 10 hr UT. See Figure 6.1 caption forr a complete description. References References ii 23

!! i | i i i i | i i i i [ ! ! j f-i i 11 ! 1 | ii r :: Ï * _ _ -- \ \ _ _ "" - - . -- .. 6 "" - - -J. : :

.. B —— r 1 1 :: ""

.. 2 2138-4533 28Aug90 UT T ii i 1 i i i i 1 i i i i I i i i > 1 i ii i i 1 ;; . 77 8 9 TIMEE CHRS) -I-I—i—[—i—i—I—rn—i—i—r —i—[—i—i—I—rn—i—i—r

B

2138-4533 29Aug90 UT ii 1 i ' i i I i i i i I i i i i I i i i ' i i 77 8 3 TIMEE CHRS)

Fig.. 6.15.— Lightcurves of the star 2138 - 453 in (a) V and (b) B obtained on sue nightss in August, 1990. See Figure 6.1 caption for a complete description. 1244 CCD Time-Resolved Photometry of Faint CVs