How to turn plastic waste into diesel fuel cheaply

[Excalibur]

Rozier56, I agree your diesel has some volatile fractions in it. A couple of things might help.

You could try raising the reflux temperature however there will be a limit because the diesel (from plastic) may become waxy when cool.

The other likely thing is to hold the diesel at say 80*C- 100*C and allow those volatile fractions to evaporate off. You would do this on the fly allowing the volatile fractions to proceed downstream to a gasoline condenser trap. Logically, it is a time vs temperature situation. A lower temp will require a longer time, while higher requires shorter.

I hope this helps.

[rozier56]

Thks Beyound,do you suggest that my reflux column is to efficient or should be shorter to allow easier flow through to the condenser stage.As it was my belief that diesel is made before petrol/gasoline.

[Beyond Biodiesel]

My new video featuring my Mk6 Turk-Retort test shows in particular the viewing port sight glass in operation. I made 130 liters in 13 hours. Feedstock was various WMO and pump fed via the lower section reflux chamber. There was no catalyst but there was stainless swarf in the reflux. There were a few problems and things to fix and sort along the way otherwise the yield would have been much better. The fuel fed turk heat was perhaps the biggest problem with flameouts.
See the video on my blog post. Click the link below...

Good work Excalibur, great video.

Hi folks,
The diesel i have produced burns with a flame at room temp{23-25*c} but the diesel sold at our pump does not.Why and what do you think even when the Sg falls in range .84-.86.the starting material is ldpe irrigation pipe.

Your diesel, rozier56, has some gasoline and/or kerosene in it. If you want to keep your fractions separate, then improve your fractionation system. Otherwise, I I find it all too simple to just check the specific gravity of the blend and run with it. I have been burning waste oils blends with gasoline, or kerosene for almost 7 years with no problems.

[rozier56]

mtrozier

Hi folks,
The diesel i have produced burns with a flame at room temp{23-25*c} but the diesel sold at our pump does not.Why and what do you think even when the Sg falls in range .84-.86.the starting material is ldpe irrigation pipe.

[Excalibur]

My new video featuring my Mk6 Turk-Retort test shows in particular the viewing port sight glass in operation. I made 130 liters in 13 hours. Feedstock was various WMO and pump fed via the lower section reflux chamber. There was no catalyst but there was stainless swarf in the reflux. There were a few problems and things to fix and sort along the way otherwise the yield would have been much better. The fuel fed turk heat was perhaps the biggest problem with flameouts.
See the video on my blog post. Click the link below...

[Excalibur]

BB. It's complex thing this Arduino PID. Actually, just thinking about the set up, I think it is probably the arduino that handles the PID and the Java based graphical front-end interface is the program for monitoring and adjustment.
It makes adjustment much, much easier. I suppose the alternative is to edit the Arduino sketch each time any change is required, then upload.
I had a very good run yesterday with the Arduino and the Front-end program. It ran without hiccup for over 12 hours

[Beyond Biodiesel]

Thanks, Excalibur, I plan to my Arduino for full PID control. I also plan to be able to monitor and download data from it as I need to. This is why I ordered a ethernet card for it, so that I can just put it on my hub. It also means that I could run it remotely. So, if the Arduino cannot do full PID control, then it is next to useless.

[Excalibur]

but I am not sure why you would need your computer involved in the loop.

The computer interface is required to control and adjust the setpoint, so for example the temperature can be lifted in steps. Also it gives a live graphic display of the 3 variables. Currently the PC is handling the PID control as well.
Even if the Arduino could do PID, at some stage an interface is required.
Looking at the setup, you'd think it's complicated and there was lots to go wrong. It's software after all.
I have the option of manual over-ride, just in case.

[Beyond Biodiesel]

Tomorrow, I'm hoping to take some video and put it up on Youtube. In particular I want to capture the diesel coming through the viewing port sight glass. I can't tell how satisfying it is to see the flow of fuel after all the hard work!! See my blog for progress.

If you have two view ports on opposite sides of the flow, then you could illuminate the flow from the rear, and photo from the front.

BB. If I can iron the bugs out of the Arduino, it will be a good step forward. I think that the advent of the very affordable microcontroller is allowing DIY's to achieve new heights. Imagine trying to do this task manually.

I agree that the Ardiuino, and other affordable microcontroller platforms, could, and have, allowed the DIY person to do excellent work, but I am not sure why you would need your computer involved in the loop. The Ardiuino should be in control, and dumping a data stream, where your computer could collect it via USB, or ethernet, or an LCD display. I plan to use all three output options with my Arduino, when I get to working on that aspect of the project. Right now I am focused upon the long range development of a more flexible and robust pyrolysis unit that can distill WMO, or crack plastic, or extract volatiles from wood, etc.; do it in a mobile package that can be operated on a daily basis.

[Excalibur]

Good news, it works OK on my quadcore PC!! Java.exe uses about 25% of the cpu and appears to be quite stable.

Ok, so tomorrow after a small welding job then a good clean up, I'll try a run of 200liters WMO.

The 80liters diesel I recently made with the retort from WMO is proving good. There's a smoothness to the way the motor runs and the economy is great.

Tomorrow, I'm hoping to take some video and put it up on Youtube. In particular I want to capture the diesel coming through the viewing port sight glass. I can't tell how satisfying it is to see the flow of fuel after all the hard work!! See my blog for progress.

BB. If I can iron the bugs out of the Arduino, it will be a good step forward. I think that the advent of the very affordable microcontroller is allowing DIY's to achieve new heights. Imagine trying to do this task manually.

[Beyond Biodiesel]

Thanks, Excalibur, for keeping us posted on your developments. The thing that this thread lacks is ardunio development for pyrolysis control.

I am still focuses upon inexpensive 800F (425c) insulation. I am very pleased with my current firebrick formula and production methods. I now have about 40-50 of them, but I need about 80 more to insulate the complete project.

I need crushed glass for the formula, and that right now is the hard labor portion of this project. Over the weekend I borrowed an electric cement mixer to churn bottles into crushed glass, but after about 8 hrs of churning I got very little crushed glass, but cleaned out my neighbor's cement mixer really well. He will be happy. I plan to go back to pounding the glass with a steel rod, because it seems to work best.

[Excalibur]

Congratulations, Excalibur, on the recent run of your WMO pyrolysis unit. I am not sure why your programming problem is a .exe issue, because I would think that since you are using an arduino, then all you need to do is down load some code into the arduino, and execute on that platform?

The microcontroller is an Arduino and the PID handling uses a Java platform. I've identified that msvcr.dll is the offending Java thread that soaks up the memory. There's numerous hits on Google about the issue but no apparent solution. The XP PC has an older single core cpu with 750mb of memory. Perhaps the old PC simply lacks resources, so to isolate the possibilty I'll substitute my quad core W7 PC with 6gb memory. I'll know by the end of today...

[Beyond Biodiesel]

Congratulations, Excalibur, on the recent run of your WMO pyrolysis unit. I am not sure why your programming problem is a .exe issue, because I would think that since you are using an arduino, then all you need to do is down load some code into the arduino, and execute on that platform?

[Excalibur]

Those of you who have been keeping an eye on my blog will know that I've done an initial run of about 80liters from WMO feedstock. I learnt a lot and have a good idea of the improvements I need to pursue.
Only when I open up the retort hatch will I be adding PE, otherwise feedstock will be continuous feed wmo.

One overwhelming point was, I need to have a clear indication of how much is in the retort at any one time. One solution is to have sight gauges on the oil-feed and diesel reservoir vessels. I can then match the input to output with perhaps an adjustment for losses.
The PID/servo software gave some grief as well. In short java.exe uses 100%cpu caused by msvcr71.dll. This eventually leads to a program or OS freeze. Should anyone know a remedy, please advise. The offending PC is XP based so I'm going to try Windows7 as a possible workaround.

Look for a Youtube video update soon.

[Beyond Biodiesel]

where are all senior members . I miss jetijs imbd excaliber asad and several others We all built our own pilot plants and sucessfully operate the same . I think this discussion is saturated now . If all senior members are reading this may I suggest to start this discussion in the same field but for scaling up this technology for bigger capacities and towards commercialisition
THANKS

Hello sunilkm153, most everyone has trouble just getting a small-scale pilot plant operational on a shoe string. I expect most of the people who made their own units found out it was a lot of trouble for too little return on the 55gallon (200L) batch scale. So, scaling it up could make it profitable, but the expense of doing so safely is often time out of the economic reach of the DIY.

Therefore my focus is distilling WMO, because it is 5 gallons (20L) in equals 5 gallons )20L) out in a 6 gallon (24L) retort. Still I have been in development for about 1.5 years so far and I still do not have a device that I can just dump 5-gallons of WMO in and get 5 gallons of fuel-grade hydrocarbons out in a few hours.

My whole pyrolysis system has been torn about for 6 months for a major rebuild, which is likely to take another 6 months, but by spring I expect to have a unit that I can do back-to-back 5-gallon batches in.