Study of the U-25B MHD generator system in strong electric and magnetic fields Page: 3 of 15
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
Figure 3 vr.jrizc', the cvp.-ri- yrtal values
of the friction coefficient, C., ,:t t". 'poratuies
of 705G and 9Jhr'<, us. a function i-eyriold", rm.-o.-r
for Toots 3 and 4. Turbulent f!,,.•» fr iction Coef-
ficients in tubes are also site,.,:, in foe figure for’
values of relative roughness fro- C.OGs to 0.1.
It is chservi"! that the average friction coef-
ficient for Test 4 was approxir Italy 0.0? and ti/.-c,
in general, friction coefficients obtained during
the fourth test are so-icwhat higher than those in
the previous test-.. This is pro:,ably due to en
increase in the effective wall roughness in Test
4 due t.o loss of sore reramc rrnterial on the
electrodes. Based on the calculat ions for tur-
bulent flow in tul.es arid on an av-'a .hydraulic
diameter for tin: channel, the effective wall rou ;h-
r.ess is estimated to he about i t 0.CG5 m.
Analysis of the will static pressure distri-
bution indicates that at a Le: persfur e of 70°C,
the channel flow was subsonic tire range of
mass flows studied, however, the temperature
was increased to Onli'L, supersonic flow existed in
the upstream section of the channel for mass flows
oT about 9 kg/s and greater, i fr-ure 4 shows the
experimental static pressure distribution along
the channel without the magnetic field for ft =
2.5 kg/s and T - 1141 K. Numbers adjacent to
the symbols designate the orifice location. A
two-dimensional an ilytical mode (l.ef. G) was u-e-d
to predict tire pressure distribution for this
case. From the previous discussion of the skin
friction coefficient, the effective v.sall roughness
has been assumed to be V. •- 0.00b si. Because of
the relatively large discrepancies between the
measured pressures on opposing walls of the
channel (lb, and i' , see rig. 4} near trie exit
of the constant-area section at tie: entrance to
the channel, two computations were periorised.
The solid curve shown in Fig. 4 gives the predicted
pressure distribution if the calculation starts
with the measured value, P_. Trie dashed curve
corresponds to ti® computed pressure distribution
where the exit pressure is kept to same as the
measured values, P,„. Figure 4 indicated that
there is a good agreement between the analytical
results arid the experimental da!a, thus giving
confidence in the measured friction coefficients.
The pressure distribution data also illus-
trate that with the combustor in operation,
subsonic flow is achieved in the channel when
the mass flow was reduced to about 1.9 kg/s
(including an additional 0.2 kg/s of" cold air
entering the combustion chamber aroung the optical
ports and through the seed nozzles). At all other
test conditions, a region of supersonic flow
existed in the upstream portion of the channel and
through a shockwave system, subsonic flow occurred
in the downs cream portion. With increase in mass
flow, the location of the shock system moved
further downstream in the channel, as discussed
in a later section.
In the upstream portion of the channel where
the flow is supersonic, the wall pressure dis-
tribution for Tests 3 and 4 were in satisfactory
agreement. In Test 4, the air gap (see Mg. 1)
located between the diffuser exit and the com-
bustion products cooling system was decreased
from 50 mm to 30 mm. This resulted in a lower
exhaust pressure than that of Test 3. Conse-
quently, the pressure distribution in the down-
stream, or subsonic, portion of the channel for
Test 4 was shifted to somewhat lower pressures
tor pared v/i th t! at of Test 3. Because of the
variations in pressure reasure; erits along the
length t; the ci.sesel , both ir- :,ts 3 err! 4,
gasdy a- is p-r a: -.-tors alor-- . channel needed
for o-.hr calculations • . obtained from averaged
valu-s o: the f./peri”-„ial static pressure dis-
tribution along the channel. A typical distributi
Of trie gaudyna-ic pars; tt*rs for tests wi i' oesi,-“(
and pota'.sius seed is summarized ir, Tab’ - ;.
Methods os-.-d to calculate the terpurature variolic
and velocity variation through the channel are dis
cusses1 in detail in i:ef. 2.
Charigo• 1 t.Metrical Perfornahce
Typical current distributions over the currer,
collect!: g electrodes measures during Test 3 are
shown in Fig. 5. The loading scheme schematic
given ir: Fig. 2 illustrates the region of current
take-off for this test. Lxtension of tbs current
collectin'; region into th-'- electrical active por-
tion of C‘e channel for Test 3 resulted ir a
somewhat smoother current distribution than that
obtain'd during Tehl 2 and described in ief. G.
Figures or: and 5b illustrate that the current
gradic-t along the take-off section boc-.e e steeper
when the oxygen enrich, ant was increased from 40'
to 00h by volume. This effect is due to the in-
crease In power by a factor of about thr.ro. Ir,
the do/.'".tre;:.;: section, currents were unifor-ly
distrib,!-••!, and increasing the oxygen enrich--/nt
merely increased the current level. Figur-s be
and jd are the current distribution at near
the max'bs' power point with that cbtaired at near
Short-C Ir.'ii t condition for a ; <ss flow of 4 kg/s
and wit;, a : agnetic field of 5 1. These current
distributer,,- s were si; liar in shape, but the
current level is shifted to higher value: as
short-circuit conditions are approach'd, for
Test 4, the current ta'.e-pff configuration was
slightly : joined as illustrated in Fig. 2. To
reduce tne possibility of flow separation ir, the
supersonic diverging portion of the channel during
Test 4, the upstream, current teke-off region was
shifted forward by 0.03 n, from the location used
in Test 3 (that is, by three frames) and into a
region oT lower magnetic field.
Typical current distributions over the take-
off regions between frames b end It, and 132 end
143 for Test 4 are shown in Fig. 6. Comparing
Figs. 5 and 6, it is noted that this shift of the
current take of," section away from the active pert
of the channel resulted in a more non-uniform
current distribution over the current collection
section. The current gradients are especially
pronounced at the lower levels of oxygen enrich-
ment, corresponding to reduced levels of plasma
conductivity.
Comparison of Figs. Ga and 6b illustrates
that a mass flow of 3 kg/s and an oxygen enrich-
ment of 50S by volume, the shape of the current
distribution over the take-off sections are
similar between operation at near maximum power
and at near short-circuit during Test 4. Only the
current level increases as the short circuit
operation condition is approached. The trend is
identical to that observed in Test 3.
Comparison of Figures 5b and Ga, for example,
which correspond to operation at near maximum
power at roughly the same test conditions clearly
shows that the upstream current distribution for
the fourth test was much steeper in the upstream
.current collection section. Comparison of Figs.
2
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Iserov, A. D.; Maksimenko, V. I. & Maslennikov, G. I. Study of the U-25B MHD generator system in strong electric and magnetic fields, article, January 1, 1979; Illinois. (https://digital.library.unt.edu/ark:/67531/metadc1097811/m1/3/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.