Flow fluctuation in stripper feed heater having 2 pass

Because of flow flcutuation in stripper feed heater having 2 pass in our lube base oil production facility, we improvised HP velocity steam line at the inlet of each pass. After injecting HP velocity steam around 2.0 ~ 2.5 t/h to the each pass inlet, pass fluctuation was disappeared but with increased inlet pressure which limit our operating flexibility . So, we want to solve this problem by modifying heater itself, but do not know the exact reason of fluctuation yet. So, anybody who experienced this kind of flow fluctution problem in heater please let me know some information. For reference, find brief operating conditions below.

Furnace inlet pressure/Temp. ; 9.5 Kg/cm2 / 205'C
inlet pipe size ; each pass has 6 inch
tube length ; around 700m for each pass
tube size ; aroud 48% of total length are
6" and the others are 5"
outlet pressure/temp.; 3.0 Kg/cm2 / 305'C
simulated vapor portion at the tube outlet is 8%
Total Charge rate to heater ; 22,000 BPD
Feed Properties ; it is the MP separator liquid coming from
reactor effluent of MSDW cat. treating
fractionator BTM of VGO HYC.
Others ; Each pass inlet has no control valves.

*As a countermeasure, we consider cut off some 5" pipes because we consider line length looks too long and installing C/V at the pass inlet with the operation of velocity steam cut.

ANYBODY WHO KNOWS THE SOLUTION BETTER THAN ME, PLEASE LET ME KNOW.
ANYBODY WHO KNOWS EXACT REASON AND POSITION WHERE FLOW FLUCTUATION STARTS IN THE TUBES, PLEASE LET ME KNOW.
ALSO, ANY INFORMATION WOULD BE VHIGHLY APPRECIATED.

Dear 1906

We have four pass fired heater on our crude unit. Flow is control valve controlled for each pass.
When we have clean heat exchangers our heater inlet pressure is 20barg, after few years in operation hater pressure is lower even 12 barg, because of heat exchanger fouling. Heater inlet temperature is 240degC.

When we have 12 barg inlet pressure, one or two passes starts to fluctuate, control valve which is on auto mode tries to control fluctuations and situations becomes worse, only solution is to put flow controller on manual. I suppose that we have small percent of feed (crude oil) vaporized on fired heater inlet. Fluctuations are in range 50m3/h, according to flow meter. With controller on manual we didn’t have problems with pass outlet temperature fluctuation..

Only cure for this situation is to try to raise pressure at heater inlet.

Can you tell us range of oscillations and frequency?
Do you have orifice flow meters on pass inlet?
Do you have problem with pass outlet temperature control?
Can you try increase pressure at heater pass inlet without steam injection?

Thanks for your specific answer. Re your requesting please find folloings.
a)Can you tell us range of oscillations and frequency?
==> When we cut the HP velocity steam, oscillations start almost immediately and the period is around 0.3~0.8 hr. And I can not tell the range in flow because we don't have flowmeter in the pass but temp. differences between passes are around 100'C
b) Do you have orifice flow meters on pass inlet?
==> As mentioned before no flowmeter we have.
c) Do you have problem with pass outlet temperature control?
==> When we have oscillation we can not control outlet temp. and the range of outlet temp. fluctuation is aroud 30 ~ 50'C
d) Can you try increase pressure at heater pass inlet without steam injection?
==> Till now no specific ways are arouse to me.

However, for your reference consider our history below ;
We have had opposite phenomena about pressure. In the past, to avoid mentioned fluctuation we decreased heater outlet stripper pressure and we got the trends of mitigating of oscillation. Any way thanks for your answer and hope deeper discussion in this forum.

Velocity steam or condensate are sometimes injected to a heater to protect the tubes from thermal degradation of the fluid and a harmful buildup of deposits. Phase separation occurs mainly because of density differences leading to an annular two-phase flow condition.

Actual flow r[é]gimes may cross from stratified to wave+mist to mist-annular. Flow maps abound.

Lack of stability in parallel tubes may be caused by a number of reasons: a changing subcooled feed condition, a vertical connection to the transfer line, uneven heat distribution with changing flow r[é]gimes, a low circulation flow rate, unequal flow rates.

Sometimes stability is regained by installing orifices at the inlets of parallel coils.
As a general rule it is recommended not to have bends in the plane of parallel coils for at least 20 diameters before the splitting point.

In this case it appears vaporization is taking place in the last half the tubes' length. I wonder whether it would be practical to inject velocity steam to the last portion, say at the start of the convection section, thus maintaining flow stability with less pressure drop.