How to turn plastic waste into diesel fuel cheaply

[jonathan]

pyrolysis oil

l try my pyrolysis again l have trouble with burner sometimes nozzle block. after 45 minutes of heating bubbles stops but begun the first drops after 15 minutes l find little fuel. l dont know the tempreture. l use waste oil about 30 litres.why bubbles stop and the fuel begun coming out? it is normal? l dont have catalyst.just empty reflux.fuel burn easy maybe diesel and petrol

[fox32]

Hello Samot, post some pictures so we can better understand your problem. DO you have a reflux? it sounds like boil-over, but i can't be sure, some pictures will help.
All the best!

[Samott]

Hi!
I use waste oil, with pp. I get one forte liter fuel liquid, then installation rough-hewn vibration so I stopped. The end I get so far a bit fuel in the first condenser but it was plastic granulous. In the second condenser was “clear” fuel. Now the first condenser it does be full of plastic (: Why?

[Beyond Biodiesel]

Yes, sounds like a good plan.
When the retort boils dry the temperature readings of any thermocouple located in the path of vapour flow will reduce significantly, right?

Correct

What readings do you get at this stage??

The temperature readings go up and down based upon what fraction is boiling; however, after the 425c peak, then the temperature will continue to decline. One just has to keep an eye on the temperatures and not allow the retort temperature to drop more than 25c before admitting more feed.

If you've been getting readings of 430C during distillation/cracking and the readings drop to 200C or something when the retort is dry we could use this as the level sensor information.

Using temperature as an indicator of a dry retort is a good idea, but you would not want it to drop that much.

If our feedstock has been pre-heated to 120C we can fill our retort until the high level TC reads 120C. During distillation/cracking we will get readings of around 430C until all volatiles are driven off and the retort is dry, then readings will drop to 200C or whatever that dry temperature is. Then we go again.

I would not do it that way. I would find a small feed rate that just keeps the retort producing distillate at a given rate.

Is there an advantage to a continuous drip feed system over this repeated batch system? I think the repeated batch system will be easy to master, yet it does expose the retort to repeated changes in temperature which may fatigue the steel, is a 300C change in temperature over and over enough to fatigue mild steel?

What you are proposing is a repeat batch system, and I think it is reasonable, until you can get the continuous drip feed system to work. As long as you have reasonable metal thickness, then metal fatigue is not likely to post an issue.

One challenge to beat on the continuous drip feed system is how to determine the high level and avoid over-filling of the retort. A retort filled to the right level, but with only the heaviest fraction remaining will still have low vapour flow rate. At least with a batch system, filling with a lower temperature feedstock means we can use a TC to tell us when the high level is reached and vapour flow rate won't confuse the issue. Though the idea of a nicely balanced drip feed system is very appealing.

Col

Well, this is a probably for the repeated batch system. A drip feed system is never going to fill the retort. All it is going to do is provide a stream of feed that flashes into vapor instantly at its maximum efficiency.

[Col]

Level sensing attempts

For a continuous feed system I am more inclined to keep the boiler nearly empty and just regulate the feed based upon out put, such that if the feed rate is too low then the output will be low; however, if the feed rate is too great then the output will be low again, so there will be some sweet spot in between, where the output is maximized. Finding that sweet spot will take some fiddling and close monitoring.

Yes, sounds like a good plan.
When the retort boils dry the temperature readings of any thermocouple located in the path of vapour flow will reduce significantly, right? What readings do you get at this stage??

If you've been getting readings of 430C during distillation/cracking and the readings drop to 200C or something when the retort is dry we could use this as the level sensor information.

If our feedstock has been pre-heated to 120C we can fill our retort until the high level TC reads 120C. During distillation/cracking we will get readings of around 430C until all volatiles are driven off and the retort is dry, then readings will drop to 200C or whatever that dry temperature is. Then we go again.

Is there an advantage to a continuous drip feed system over this repeated batch system? I think the repeated batch system will be easy to master, yet it does expose the retort to repeated changes in temperature which may fatigue the steel, is a 300C change in temperature over and over enough to fatigue mild steel?

One challenge to beat on the continuous drip feed system is how to determine the high level and avoid over-filling of the retort. A retort filled to the right level, but with only the heaviest fraction remaining will still have low vapour flow rate. At least with a batch system, filling with a lower temperature feedstock means we can use a TC to tell us when the high level is reached and vapour flow rate won't confuse the issue. Though the idea of a nicely balanced drip feed system is very appealing.

Col

[Beyond Biodiesel]

I am not sure, Col, whether the probe will work, because as you say, the bubbles might confuse it. So, if you buy one, then let us know how it goes.

For a continuous feed system I am more inclined to keep the boiler nearly empty and just regulate the feed based upon out put, such that if the feed rate is too low then the output will be low; however, if the feed rate is too great then the output will be low again, so there will be some sweet spot in between, where the output is maximized. Finding that sweet spot will take some fiddling and close monitoring.

[Col]

high temperature liquid level probe

Hi guys.
Knowing / maintaining the liquid level in the retort is important for any continuous plant. I am considering this ceramic level sensor probe capable of withstanding 800C. Price is $250.
DRFK-98 RF Admittance Level Switch, View fuel level switch, Dongrun Product Details from Yantai Dongrun Instrument Co., Ltd. on Alibaba.com
Has anyone had any experience with similar level sensing probes in this or other applications where the tank is bubbling / frothing? Do they sense the level of the liquid or the level of the bubbles / froth?
For my small retort I am thinking I'll make it from a piece of 270mm diameter vertical pipe, intersecting another horizontal piece (making a 'T'), and keep the oil level at the half full mark for the horizontal piece. 200mm vertical, 500mm long horizontal. This gives a 25L capacity retort to the half full mark on the horizontal piece. The 270mm diameter pipe will suit a high pressure LPG burner I will use initially while I iron out the bugs before converting to a larger naptha burner later. I think the half empty horizontal section should serve as an expansion vessel and 'bubble-popper', increasing the available room and surface area for any boil-over to settle down.
I also think that even if the probes are sensitive to bubbles, the bubbles should be able to be excluded from the probe by some baffles so only the liquid is sensed. Any feedback anyone?
Col

[Col]

Flame propagation speed is a known issue.
Some saved links on the subject. Hope it helps:

Excellent, thank you Excalibur.
Col

[Excalibur]

In relation to the flame arrestor - does anyone think there would be a problem if the diameter of the pipe stuffed with steel wool was larger than required by the system? I want to build the system downstream from the retort oversized so that I can simply swap out the retort but keep the rest when I am ready to upscale. I think an oversized flame arrestor will let oxygen / air pass back up to the pipe to the bubbler but it would stop there so I can't see it being a problem in a normal operational scenario. But if the flame travelled back to the steel wool, and there was a lot of steel wool in a large diameter pipe I wonder if the flame could still maintain enough velocity to pass through the steel wool. Or perhaps a lot of steel wool is even better because the heat quench will be better. I think I'll submerge my flame arrestor in water just to be sure (not the end, just the middle part). I want to have 100mm diameter pipes running from the retort to the last condenser, then 50mm steel pipes from the bubbler to the flame arrestor. But at first the system might only need a 20mm flame arrestor. Any thoughts?
Col

Complex subject. Has been a while since I read up on it. Flame propagation speed is a known issue. One link below had a calculation re pipe size vs distance. Keep the arrestor cool. Obviously we are already excluding oxygen. Some saved links on the subject. Hope it helps:
Kemp PDF

About Flame Arrestors and Detonation Arrestors

Flame arresters certified to EN ISO 16825/12874/ATEX - Flammer GmbH

I would like to ask for measure of reaktor as I need pass them to welder man. I need diameter and hight of vessel.

I don't recall Jetijs quoting exact dimensions so make it to suit available materials aimed at the size of reactor you want.

[Beyond Biodiesel]

Dear Mr. Beyond Biodiesel Thank you for your interest, but is it reasonable to use the retort area of 100 liters to put inside an empty 10 kg of plastic to produce 10 liters of fuel? And what about you How much space you have and how much the reactor put inside plastic? Wait for your reply eagerly repeat my thanks to you, but Google translation

Correct, theoretically it takes 100L of plastic to make 10L of cracked hydrocarbon liquid.

My largest retort is only about 40L, but I am working with WMO, not plastics for now. When I move to plastics I plan to build a 220L retort.

From what i understand i reckon you want to know how much free space it reamains? If this is the case, you have to leave at least a quarter (1/4) from your reactor free, to avoid any boil over. You cannot fill it up entirely. Hope this helps.

fox32 is correct, dedooo, so if your reflux is going to simply be the upper 1/4 of your retort, then you will need 125L retort with 100L of shredded plastics to crack into 10L of liquid hydrocarbons. However, if you have a separate reflux , then you need not do that.

In relation to the flame arrestor - does anyone think there would be a problem if the diameter of the pipe stuffed with steel wool was larger than required by the system? I want to build the system downstream from the retort oversized so that I can simply swap out the retort but keep the rest when I am ready to upscale. I think an oversized flame arrestor will let oxygen / air pass back up to the pipe to the bubbler but it would stop there so I can't see it being a problem in a normal operational scenario. But if the flame travelled back to the steel wool, and there was a lot of steel wool in a large diameter pipe I wonder if the flame could still maintain enough velocity to pass through the steel wool. Or perhaps a lot of steel wool is even better because the heat quench will be better. I think I'll submerge my flame arrestor in water just to be sure (not the end, just the middle part). I want to have 100mm diameter pipes running from the retort to the last condenser, then 50mm steel pipes from the bubbler to the flame arrestor. But at first the system might only need a 20mm flame arrestor. Any thoughts?
Col

I see no problem with enlarging the section after the retort, then replacing the retort later with an upgrade, as long as the enlarged pipe sections are no more than 4x the diameter of the retort exhaust section, otherwise you could end up with a non-positive pressure situation that could bring in air, but your bubble still functions as a barrier to oxygen.

Presumably you have some kind of flame at the exhaust exit that will burn flammables? There is an advantage here in using a flame on the retort for heat, so that you can direct the exhaust gasses there to consume them, and reduce your need for additional fuel to run you retort.

[peterkyjovsky]

Driwings

Hi all.
I made a 3d cad drawing of the reactor, lid, cage, brick ring and outer barrel assembly. You will need edrawings software to view the file, you can download it for free here:
eDrawings - 3D and 2D CAD Viewer and Publisher

After that you can view the assembly in 3d and make parts transparent to see what is behind them. Hope this will help to understand the design better. The hollow part between the brick rings and outer barren should be stuffed with glassfirber insulation. Here is the drawing file:

pyrolysis reactor assembly.EASM

I would like to ask for measure of reaktor as I need pass them to welder man. I need diameter and hight of vessel.

[fox32]

Dear Mr. Beyond Biodiesel Thank you for your interest, but is it reasonable to use the retort area of 100 liters to put inside an empty 10 kg of plastic to produce 10 liters of fuel? And what about you How much space you have and how much the reactor put inside plastic? Wait for your reply eagerly repeat my thanks to you, but Google translation

From what i understand i reckon you want to know how much free space it reamains? If this is the case, you have to leave at least a quarter (1/4) from your reactor free, to avoid any boil over. You cannot fill it up entirely. Hope this helps.

[Col]

Flame arrestor

In relation to the flame arrestor - does anyone think there would be a problem if the diameter of the pipe stuffed with steel wool was larger than required by the system? I want to build the system downstream from the retort oversized so that I can simply swap out the retort but keep the rest when I am ready to upscale. I think an oversized flame arrestor will let oxygen / air pass back up to the pipe to the bubbler but it would stop there so I can't see it being a problem in a normal operational scenario. But if the flame travelled back to the steel wool, and there was a lot of steel wool in a large diameter pipe I wonder if the flame could still maintain enough velocity to pass through the steel wool. Or perhaps a lot of steel wool is even better because the heat quench will be better. I think I'll submerge my flame arrestor in water just to be sure (not the end, just the middle part). I want to have 100mm diameter pipes running from the retort to the last condenser, then 50mm steel pipes from the bubbler to the flame arrestor. But at first the system might only need a 20mm flame arrestor. Any thoughts?
Col

[dedooo]

Dear Mr. Beyond Biodiesel Thank you for your interest, but is it reasonable to use the retort area of 100 liters to put inside an empty 10 kg of plastic to produce 10 liters of fuel? And what about you How much space you have and how much the reactor put inside plastic? Wait for your reply eagerly repeat my thanks to you, but Google translation

[dedooo]

Dear Mr Excalibur. looks like you did not understand me very probably due to the translation of Google, will rephrase my question: How is the area of the reactor iron needed to produce 10 liters of fuel, I know that 10 kg of plastic gives about 10 liters of fuel / but Reactor in my area of 29 liters, a propane gas cylinder empty house, widened to only 5 or 7 kg of waste plastic