A long nededle verus a short one

[smott2011]

I'm a vascular access nurse for a Kindey dialysis center my question is this; when we cannulated with a 1 inch needle and run the pump at 400 we get art and venous pressures of 240 to 280. When we use the 3/5 inch needles the pressure drops to 160 to 180 what is the law of physics for this in novice terms
Stuart Mo

 

[JaredJames]

I wouldn't say the length has much bearing on this, you'll probably find the internal diameter of the needle is different.

Giving you are pumping at a constant pressure, the smaller the internal diameter of the needle, the faster the flow through it and so the lower the pressure.

It's the venturi effect.

 

[Rap]

I'm a vascular access nurse for a Kindey dialysis center my question is this; when we cannulated with a 1 inch needle and run the pump at 400 we get art and venous pressures of 240 to 280. When we use the 3/5 inch needles the pressure drops to 160 to 180 what is the law of physics for this in novice terms
Stuart Mo

I think the Poiseuille effect is what's needed here - there is friction between the fluid and the walls of the needle.

(Pi-Po)=\frac{8\mu L Q}{\pi r^4}

where Pi and Po are the input and output pressures, L is the length of the needle, Q is the volume flow rate, r is the needle radius, and \mu is the viscosity.

If I understand what you are saying, Pi=400, Po=260 for the 1 inch needle, Po=170 for the 0.6 inch needle. So if \mu,Q, and r stayed constant, (Pi-Po)/L should stay constant, but it does not: (400-260)/1 is much smaller than (400-170/0.6. That means that something else is changing, the radius of the needle or the flow rate thru the needle (not the viscosity).

Do you know the radius and the flow rate for the two needles?

 

[JaredJames]

I think the Poiseuille effect is what's needed here - there is friction between the fluid and the walls of the needle.

(Pi-Po)=\frac{8\mu L Q}{\pi r^4}

where Pi and Po are the input and output pressures, L is the length of the needle, Q is the volume flow rate, r is the needle radius, and \mu is the viscosity.

If I understand what you are saying, Pi=400, Po=260 for the 1 inch needle, Po=170 for the 0.6 inch needle. So if \mu,Q, and r stayed constant, (Pi-Po)/L should stay constant, but it does not (400-260)/1 is much smaller than (400-170/0.6. That means that something else is changing, the radius of the needle or the flow rate thru the needle (not the viscosity).

Do you know the radius and the flow rate for the two needles?

The friction difference over such a small length is negligible. You're talking of 2/5 of an inch.

If all values of your above are constant aside from Po and L, the change in L doesn't make up for the change in Po (a conclusion you draw yourself).

The flow rate will be determined by the diameter of the needle.

So I stand by the needles internal diameter conclusion.

 

[Rap]

The friction difference over such a small length is negligible. You're talking of 2/5 of an inch.

3/5 of an inch, but point taken.

Well, we both agree the radius has an effect. I have not checked the pressure drop we might expect from the Poiseuille effect, but if you know it is small, I will take your word for it. But if there is no friction in the needle, then there is no other force on the fluid while it is in the needle, and that means there is no pressure drop across the needle.

That would mean the pressure drop occurs before (or at the entrance to) the needle, yet is caused by the needle. I agree, the Venturi effect covers that, the Venturi effect says

(Pi-Po)=\frac{\rho}{2}(v_i^2-v_o^2)=\frac{\rho Q^2}{2}(1/A_i^2-1/A_o^2)

where Ai is the larger input cross sectional area and Ao is the cross sectional area of the needle, and Q is again the volumetric flow rate.

So if we could get the radii and flow rates for the two needles, we would know a lot more about what's going on. So again I ask Smott2011 if the radii and flow rates are availiable for the two cases.

 

[JaredJames]

3/5 of an inch, but point taken.

You have a 1 inch needle (5/5 inch) and a 3/5 inch needle - the difference is 2/5 inch.

Well, we both agree the radius has an effect. I have not checked the pressure drop we might expect from the Poiseuille effect, but if you know it is small, I will take your word for it. But if there is no friction in the needle, then there is no other force on the fluid while it is in the needle, and that means there is no pressure drop across the needle.

That would mean the pressure drop occurs before the needle, yet is caused by the needle. I agree, the Venturi effect covers that, but the Venturi effect says (using Q as volume flow rate again).

So if we could get the radii and flow rates for the two needles, we would know a lot more about what's going on. So again I ask Smott2011 if the radii and flow rates are availiable for the two cases.

We don't need the flow rates, we can calculate them with entry/exit pressures and needle dimensions.

The pressure drop occurs within the needle due to the flow rate increase. The smaller the diameter the greater the flow rate inside and the lower the pressure. That's really all there is to it.

 

[Rap]

You have a 1 inch needle (5/5 inch) and a 3/5 inch needle - the difference is 2/5 inch. We don't need the flow rates, we can calculate them with entry/exit pressures and needle dimensions.

Yes, as long as we are sure of our model.

The pressure drop occurs within the needle due to the flow rate increase. The smaller the diameter the greater the flow rate inside and the lower the pressure. That's really all there is to it.

For the Venturi effect, I think the pressure drop occurs at the interface where the radii change. Unless there is an opposing force along the length of the needle (i.e. viscous), there will be no pressure gradient along the length of the needle.

PS - sorry for the Venturi equation confusion.

 

[cjl]

The friction difference over such a small length is negligible. You're talking of 2/5 of an inch.

That depends on whether viscous or pressure/inertial forces are more important here. That is a difference of 40%, which would be pretty significant if the flow was dominated by viscosity effects. Of course, given the stated data, I would expect that the internal diameter between the two needles is significantly different, which would also need to be taken into account.

 

[smott2011]

[OK everything is the process pump speed all the same they only difference is the length.1 Inch versus 3/5 inch. I have been told that if you had a garden hose 50 feet long and one 250 long you would have a different pressures this the venturi effect. The reason i need this info is that I'm doing a article for a magazine. Where would i be able to find articles to support this . One more question would there be any cell damage the 3/5 versus the 1 inch? Human blood flows threw

Thank Yo

 

[JaredJames]

I have been told that if you had a garden hose 50 feet long and one 250 long you would have a different pressures this the venturi effect.

No, that is not the venturi effect.

The difference in pressure here is due to resistance to flow. There will be R resistance per foot so a 50 foot hose will have 50 x R and 205 foot will have 250 x R - that gives you the difference in pressures.

The reason i need this info is that I'm doing a article for a magazine. Where would i be able to find articles to support this .

You won't, as above. It's wrong.

One more question would there be any cell damage the 3/5 versus the 1 inch? Human blood flows threw

If you're writing for a magazine, note spelling and grammar = threw should be through.

This is a question to ask a biologist, not a physicist.

 

[smott2011]

The friction difference over such a small length is negligible. You're talking of 2/5 of an inch.

If all values of your above are constant aside from Po and L, the change in L doesn't make up for the change in Po (a conclusion you draw yourself).

The flow rate will be determined by the diameter of the needle.

So I stand by the needles internal diameter conclusion.

Everthing is the same diameter flow the only difference is the length 1 inch versa 3/5. we get the same results on thousands of patints[

 

[JaredJames]

Everthing is the same diameter flow the only difference is the length 1 inch versa 3/5. we get the same results on thousands of patints[

In which case, it is down to the viscosity of blood and the length is the controlling factor - assuming you haven't forgotten something.

I didn't realize how much 'thicker' blood is than water, according to the numbers it's about 4 times more viscous. So it could certainly account for it over the length.

You need to look up viscous flow and flow through pipes. Churn some numbers and you should come out with your answers.

 

[Rap]

Everthing is the same diameter flow the only difference is the length 1 inch versa 3/5. we get the same results on thousands of patints[

Do you know if there is any difference in the rate that blood flows through the needle? Is there any measurements that tell how much blood is flowing through per minute?

 

[smott2011]

The flow is always the same needle size is the same except for the length 1 inch versus 3/5 inch. so what do you all thing is the physics equations for this?

 

[Rap]

The flow is always the same needle size is the same except for the length 1 inch versus 3/5 inch. so what do you all thing is the physics equations for this?

Well, there are two effects - Venturi effect and Poiseuille effect. It would really help if you knew the diameter of the two needles (inside and outside). Is it like a hypodermic needle? Do you know what the flow rate is?. Do you have any of these numbers? Do you know how much total blood is being recirculated? Do you know how much time it takes to to do one complete recirculation? Anything will help.

 

[JaredJames]

Well, there are two effects - Venturi effect and Poiseuille effect. It would really help if you knew the diameter of the two needles (inside and outside). Is it like a hypodermic needle? Do you know what the flow rate is?. Do you have any of these numbers? Do you know how much total blood is being recirculated? Do you know how much time it takes to to do one complete recirculation? Anything will help.

I must concur.

Physics won't say anything without these numbers.

It appears you just want an equation you can show off with for the magazine, but that alone doesn't mean anything. If that's all you want, you might as well just give the stuff related to the venturi effect and be done with it.

 

[smott2011]

outer diameter 1.81 to1.85mm

 

[Rap]

outer diameter 1.81 to1.85mm

How about flow rates?

 

[smott2011]

all are 400

 

[JaredJames]

all are 400

That's a pressure not a flow rate isn't it?

 

[Rap]

all are 400

In order to do this right, we really have to get flow rates, including units. You have to follow 400 by units, like liters per second or something. And yes, 400 was the input pressure, 400 psi, that's not the flow rate.

 

[smott2011]

OK i will have that for you tomorrow

 

[smott2011]

400 hundred milliliters per minut

 

[JaredJames]

400 hundred milliliters per minut

Lovely, I don't suppose you have the pressures? (I understand you gave the outlet side, what about the inlet?)

I'd also request the internal diameter of the needles (I'm under the impression the one you gave was external?)

 

[Rap]

Lovely, I don't suppose you have the pressures? (I understand you gave the outlet side, what about the inlet?)

I'd also request the internal diameter of the needles (I'm under the impression the one you gave was external?)

I checked needles, assuming it was OD. The corresponding ID is 1.372 mm. Heres my Poiseuille calculation

\mu=3.5e-3 kg m/sec
L=2.54e-2 m (1 inch)
Q=6.66e-6 m^3/sec (400 ml/min)
r=6.86e-4 m

Using

\Delta P=\frac{8\mu L Q}{\pi r^4}

I get delta P = 6808 pa = 0.99 psi.

As I understood the input pressure was 400 psi, output pressure 260 and 170 psi for the 1 and 0.6 inch needle respectively. This tells me its not Poiseuille effect.

The problem with the Venturi effect is that it is not a function of needle length. Unless there is a third possibility, this tells me that the problem we are solving is not the problem at hand.

Check out http://pubs.acs.org/doi/abs/10.1021/es00141a021 - I do not have a subscription, but I would be interested in what it says.

 

[JaredJames]

As I understood the input pressure was 400 psi, output pressure 260 and 170 psi for the 1 and 0.6 inch needle respectively. This tells me its not Poiseuille effect.

The flow rate is 400, not the pressure (or at least I'm waiting for clarification on that).

The problem with the Venturi effect is that it is not a function of needle length. Unless there is a third possibility, this tells me that the problem we are solving is not the problem at hand.

I'd be interested to know if both lengths of needle have the same internal dimensions.

 

[Rap]

I'm a vascular access nurse for a Kindey dialysis center my question is this; when we cannulated with a 1 inch needle and run the pump at 400 we get art and venous pressures of 240 to 280. When we use the 3/5 inch needles the pressure drops to 160 to 180 what is the law of physics for this in novice terms
Stuart Mo

jarednjames - good point.

smott2011 - Please state what "240 to 280" means. Is that pressure in psi, mm of mercury?
Same for "160 to 180". Also, what exactly does "run the pump at 400" mean. 400 ml/minute? 400 psi? 400 mm of mercury?

 

[Skrambles]

If you assume the density of blood to be the same as water (1 gram per mL) and use the numbers Rap provided, that gives a Reynolds number of approximately 1800. This suggests you are probably experiencing a pressure drop due to turbulence.

There are at least two other issues you should consider.

The coupling between the needle and the tubing from the pump may cause a noticeable pressure drop. Are the couplings the same style and dimension for both needles?

Blood is not a Newtonian fluid, so its viscosity will change depending on the shear stress. It may not be significantly different between those two pressures though.

Turbulence seems the most likely cause; 400 mL/min is a pretty high flow rate for a needle of that diameter.

 

[JaredJames]

If you assume the density of blood to be the same as water (1 gram per mL) and use the numbers Rap provided, that gives a Reynolds number of approximately 1800. This suggests you are probably experiencing a pressure drop due to turbulence.

The dynamic viscosity of blood is approximately 4 to 5 times that of water. Big difference and something that needs to be factored in.

Bloods density is around 1060kg/m³ from what I've seen.

 

[Skrambles]

The dynamic viscosity of blood is approximately 4 to 5 times that of water. Big difference and something that needs to be factored in.

Bloods density is around 1060kg/m³ from what I've seen.

The number I used for viscosity was about 3.5 times that of water, but it will vary from person to person and can change due to temperature and pressure.

The density you gave is almost exactly the same as water.

The Reynolds number also depends on the diameter of the pipe, in this case the needle. A small change in this variable can have a huge effect on turbidity. Just because the needles are specified at the same inner diameter doesn't mean they are actually the same. Unfortunately that dimension is extremely hard to measure.

 

[JaredJames]

The Reynolds number also depends on the diameter of the pipe, in this case the needle. A small change in this variable can have a huge effect on turbidity.

If a small change in diameter can have a huge effect, so can a small change in density. Neither are squared / cubed (etc) so equal changes in either would have the same effect.

The density you gave is almost exactly the same as water.

The point of the above being that you are talking about microscopic changes in diameter internally having a "huge effect", and yet you have a far bigger change in density between your assumption of water and actual blood.

 

[smott2011]

the art pressure with the one inch is 240-280 Venus pressure is 240-280 the 3/5 inch is 160-180 Venus is 160-180. this is mercury pressure that shows on the dialysis machine the 400 is 400 hundred milliliters per minutes. once again let me state the only difference is the length of the needle one inch versus 3/5 inch

 

[Rap]

the art pressure with the one inch is 240-280 Venus pressure is 240-280 the 3/5 inch is 160-180 Venus is 160-180. this is mercury pressure that shows on the dialysis machine the 400 is 400 hundred milliliters per minutes. once again let me state the only difference is the length of the needle one inch versus 3/5 inch

Ok, then the result I got in #25 (6808 pa or about 1 psi) gives 51 torr (mm hg). This puts it in the Poiseuille range. I imagine the pressure 260 torr and 170 torr are above the body's pressure of maybe 120 torr? So that's 140 torr and 50 torr for the one inch and 0.6 inch, respectively. Missing a factor of three - and that could well be in the viscosity coefficient. I used 3.5 times that of water.

For the Poiseuille effect, the pressure drop is proportional to the length. Pressure over length should be the same. For the one inch its 140/1=140. For the 0.6 inch its 50/0.6=83. I would think it would be closer.

I used http://en.wikipedia.org/wiki/Needle_gauge_comparison_chart to figure the ID of both needles, assuming from #17 that OD was 1.81 to1.85mm.

 

[Skrambles]

If a small change in diameter can have a huge effect, so can a small change in density. Neither are squared / cubed (etc) so equal changes in either would have the same effect.


The point of the above being that you are talking about microscopic changes in diameter internally having a "huge effect", and yet you have a far bigger change in density between your assumption of water and actual blood.

Re = (\rho * V * d) / \mu

V is average velocity of the flow, and d is the pipe diameter.

1060 kg/m^3 = 1.06 g / mL That's a 6% difference from the number I used.

4.5 centiPoise / 3.5 centiPoise = 1.286 That's a 28% difference in viscosity from what I used.

The diameter I used, based on the radius Rap provided, is 13.72 x 10^-4 meters, or 1.372 mm. Because this number is so small, it would not take much of a change to make a large percentage difference.

The Reynolds number is a dimensionless number used to approximate the turbidity of a given flow. Also, if the flow is turbulent, then the Hagen-Poiseuille equation is not applicable. It is meant for laminar flow only.

 

[Rap]

The Reynolds number is a dimensionless number used to approximate the turbidity of a given flow. Also, if the flow is turbulent, then the Hagen-Poiseuille equation is not applicable. It is meant for laminar flow only.

Using Q=A*v where Q is the flow rate (400 ml/minute) and A is the cross sectional area of the needle, I get velocity in the needle as 4.508 m/sec. Using the expression for the Reynolds number, I get Re=1871, which is in the laminar flow regime. I don't know, but I would guess turbulence and blood would not be a good combination clot-wise.

DUH! It's probably a combination of the Poiseuille and Venturi effects.

\Delta P = \frac{\rho Q^2}{2}(1/A^2-\1/A_i^2) + \frac{8\mu L Q}{\pi r^4}

Where Ai is the cross sectional area of the tube or pipe leading to the needle, L is length of needle. Assuming Ai>>A, we can ignore 1/Ai. Using density 1060 kg/m^3, mu=3.5e-3 kg/m/sec, I get for L=1 inch Delta P = 51.06+80.67 =131.73 torr. For L=0.6 inch, I get 30.64+80.67=111.30 torr. As per #33, I would expect 140 and 50 torr. I wonder if this is within the limits of the uncertainty for our values of the various parameters.

 

[Skrambles]

The coupling between the needle and tube may be a significant factor here. I would guess that they are using a luer-lock, and if so then there would not be a nice smooth taper between the tube and needle. The needle could also be slightly occluded, which would increase the pressure needed to maintain the flow rate.

 

[JaredJames]

The Reynolds number is a dimensionless number used to approximate the turbidity of a given flow. Also, if the flow is turbulent, then the Hagen-Poiseuille equation is not applicable. It is meant for laminar flow only.

Spoken like a textbook. Now keep reading and you'll see 1800 is well inside the laminar range. Not sure how you came to the turbulent conclusion above.

Rap is correct (assuming our numbers are good).

 

[Skrambles]

Spoken like a textbook. Now keep reading and you'll see 1800 is well inside the laminar range. Not sure how you came to the turbulent conclusion above.

Rap is correct (assuming our numbers are good).

Actually, 1800 is NOT well inside the laminar range, it is in the transitional regime between fully laminar and fully turbulent flow. Perhaps it would help you to actually read a textbook before being so condescending.

It's quite funny that you're a PF Contributor considering you have contributed almost nothing to this thread.

My job is in medical device development, and I have about two dozen syringes on my desk at this very moment. I have to think about this type of problem on an almost daily basis.

If you would like to insult me some more, please do so through a private message.

 

[JaredJames]

Actually, 1800 is NOT well inside the laminar range, it is in the transitional regime between fully laminar and fully turbulent flow. Perhaps it would help you to actually read a textbook before being so condescending.

Transitional reynolds flow is approximately 2300 to 4000 (although these numbers can vary within around 200 of the stated values depending on where you look), turbulent is above 4000.

I have never seen a textbook state 1800 as being in the transitional stage.

http://www.princeton.edu/~asmits/Bicycle_web/transition.html

While the transition from laminar to turbulent flow occurs at a Reynolds number of approximately 2300 in a pipe, the precise value depends on whether any small disturbances are present. If the experiment is very carefully arranged so that the pipe is very smooth and there are no disturbances to the velocity and so on, higher values of Re can be obtained with the flow still in a laminar state. However, if Re is less than 2300, the flow will be laminar even if it is disturbed. Thus 2300 is the value the Re below which turbulence will not occur in a pipe.

As someone who works with needles (no sarcasm, completely serious), how smooth are they internally? Perhaps that has a bearing on the flow?

I completely agree with you that I feel it's the connections that cause the problem, not the needle itself. That could certainly induce turbulent flow.

 

[Skrambles]

That 2300 to 4000 range is commonly used in textbooks, but with a little digging you will find published values that are much different. Those numbers are also meant to apply only to Newtonian fluids flowing through a rigid pipe that has a length several factors greater than its inner diameter. Blood is definitely not a Newtonian fluid, and the length of the needle in these cases is not enough to meet the second requirement.

http://books.google.com/books?id=sz...page&q=critical reynolds number blood&f=false

This author puts the critical Reynolds number at 1000 for a smooth, rigid, cylindrical tube.

http://wiki.nus.edu.sg/display/Hemolysis/Turbulence

"A transition state occurs between Re values of 2000 to 4000, and above 4000 the flow is said to be turbulent. However these are general guidelines and whether the flow is turbulent or not depends on other factors."

http://ep.physoc.org/content/44/1/110.full.pdf

"The velocity of flow in the vascular system is sufficiently high for turbulence to occur and Coulter and Pappenheimer [1949] found by experiment that the critical Re for bovine blood flowing in a straight tube was between 1800 and 2100."

Here's an interesting paper on turbulence in flexible tubes:

http://chemeng.iisc.ernet.in/kumaran/publications/1999-1997/vkumaran_jfm357_1998.pdf

"An attempt was made to observe the onset of turbulence by the injection of a dye stream into the tube. The authors reported that the dye stream in the centre of the tube becomes chaotic at a Reynolds number between 570 and 870. However, the authors cautioned that this might overestimate the transition Reynolds number since the turbulence appears to originate at the walls and then grows inwards to engulf the laminar core."

The needle being used is probably connected to the pump with surgical tubing, and this may also be a source for turbulence in the system. The only way to know for sure would be through experimentation.

 

[JaredJames]

Interesting stuff.

So would the needle be the narrowest part of the system or would it's diameter not vary much from the tubing?

 

[Rap]

That 2300 to 4000 range is commonly used in textbooks, but with a little digging you will find published values that are much different. Those numbers are also meant to apply only to Newtonian fluids flowing through a rigid pipe that has a length several factors greater than its inner diameter. Blood is definitely not a Newtonian fluid, and the length of the needle in these cases is not enough to meet the second requirement.

http://books.google.com/books?id=sz...page&q=critical reynolds number blood&f=false

This author puts the critical Reynolds number at 1000 for a smooth, rigid, cylindrical tube.

http://wiki.nus.edu.sg/display/Hemolysis/Turbulence

"A transition state occurs between Re values of 2000 to 4000, and above 4000 the flow is said to be turbulent. However these are general guidelines and whether the flow is turbulent or not depends on other factors."

http://ep.physoc.org/content/44/1/110.full.pdf

"The velocity of flow in the vascular system is sufficiently high for turbulence to occur and Coulter and Pappenheimer [1949] found by experiment that the critical Re for bovine blood flowing in a straight tube was between 1800 and 2100."

Here's an interesting paper on turbulence in flexible tubes:

http://chemeng.iisc.ernet.in/kumaran/publications/1999-1997/vkumaran_jfm357_1998.pdf

"An attempt was made to observe the onset of turbulence by the injection of a dye stream into the tube. The authors reported that the dye stream in the centre of the tube becomes chaotic at a Reynolds number between 570 and 870. However, the authors cautioned that this might overestimate the transition Reynolds number since the turbulence appears to originate at the walls and then grows inwards to engulf the laminar core."

The needle being used is probably connected to the pump with surgical tubing, and this may also be a source for turbulence in the system. The only way to know for sure would be through experimentation.

Again - it seems to me that turbulence would be a bad thing when piping blood. It would make the probability of clotting much higher. Just on that basis, I would expect the design, whatever it was, would be good enough to prevent turbulence.

 

[Skrambles]

Interesting stuff.

So would the needle be the narrowest part of the system or would it's diameter not vary much from the tubing?

I don't know for sure what tubing would be used for dialysis, but I seriously doubt it would be that narrow inside.

From what I've been reading there is turbulence that occurs naturally in the vascular system, especially near the heart. I'm not a hematologist so I don't know what effects these conditions might have on blood. Frankly, I don't even like to see blood (especially my own).

 

[Rap]

So how do we determine if the flow is turbulent?