Pid controlled [motorized] pressure regulator?

Fred.Rowe

Fred.Rowe

Joined
May 2, 2004
Messages
6,848
I am in the process of building a 12 inch cylinder forge. This is the third forge I have built. It is a blown forge, using well head [natural gas] as the fuel. There is a "K" thermocouple installed inside the forge chamber.
I have manual control of the, fuel and air, both volume and pressure.
My question is this; Is it possible to operate a motorized pressure regulator, using a pid controller that is connected to the "K" thermocouple?
I would like to use the manual controls, gate valves, needle valves, to tweak the burners, but would like to be able to set the temp on the pid and let it regulate the pressure, therefore regulating the temperature inside the forge.
Has anyone taken this approach in building there forge?
On paper, it seems like a good idea. But I have seen ideas that sounded good end up not working in a practical operation.



Thank you in advance for your input, Fred
 
I think that your setup would not only have to regulate pressure, but air flow, if you tried to make it happen. Reason being because you would change the fuel mix badly if you simply changed pressure. Trying to do all of this would require a feedback system to work well, which is beyond the range of most DIY.
May be others though that know a simple solution for this kind of thing.
 
Any Cal. said:
I think that your setup would not only have to regulate pressure, but air flow, if you tried to make it happen. Reason being because you would change the fuel mix badly if you simply changed pressure. Trying to do all of this would require a feedback system to work well, which is beyond the range of most DIY.
May be others though that know a simple solution for this kind of thing.
Click to expand...

With a venturi burner, when you turn the gas pressure down, the air intake through the venturi, is reduced. I don't know if it is proportionate or not.

Not so, with a blower. The volume would remain the same, no matter the gas pressure.
It would be easy enough to add an air modulator valve and run it off the same controller. The Watlo I run now has many function modes.
It would make for a sweet setup. I'm building this one for fun, more than necessity. I can't afford my dream car so this will have to do.:thumbup::D

Thanks for your input, Fred
 
This is an interesting question, Fred, and could be a fun discussion. My somewhat jumbled thoughts:

In short, the answer of "Could it be done at all?" is certainly "yes". In one scenario, it would take a reversible stepper motor connected to a custom pressure regulator, a feedback loop to measure the pressure regulation (pressure monitor) versus the response (thermocouple), and, most importantly, a control algorithm specifically written for the system to determine how much pressure change produces the desired effect (the logic circuit, so to speak). Such systems are used in some lab equipment I am familiar with, and quite high-tech, purpose-built, and expensive.

Not having specific knowledge about it, I would nonetheless suspect that a simple off-the-shelf controller wouldn't have the type of customized logic built in for the proper control. That said, they are proportional devices, so I could be all wet. I don't think, though, that the on/off control of a typical controller could be made to work, especially since there is a lag in change/response.

Since controlling pressure through a fixed orifice is essentially controlling flow rate, perhaps it would be more appropriate to conside a proportional flow valve instead of the typical open/closed solenoid usually seen in these applications?

My concern with proportional flow of only one feed stream (gas and not air) would be that the change in mixtures would resulting in a constantly changing atmosphere, and it would be hard to anticipate when one is sudddenly going to go from reducing to neutral to oxidizing.

Just some random thoughts. Maybe it'll stimulate more discussion. There are some people around here who work with control circuits, so maybe they'll see this and comment.
 
Mike hit some good points. PID control of temperature is difficult at best. Temperature is slow, valves are fast.....relatively speaking. I worked for a number of years designing and building propane/air mixing systems to be used in place of natural gas. To achieve a constant mix at variable flows requires a specially designed dual body valve with a single valve stem. They were large systems capable of 200k scfm with a 20/1 turn down. I would say "it's possible but not practical.
Of course that was 25 years ago and technology has come a long way.

Mike Broach
 
I believe there are just too many variables, to monitor and meter, to make it a simple fix.
This is an idea that got booted along through conversations with other makers on how to build the ultimate forge.

This is a five burner forge, I'm building, with volume control for each burner. It has a 3 inch valve, to adjust the air, a needle valve for the gas along with the pressure regulator. I have around 18 lbs on my ng line.

Its hard to beat what we were born with; Two good hands and a brain to instruct them.:thumbup::cool:
 
Fred

Is there a place around there that fixes old stoves.

The older stoves (20’s - 40’s) had ovens hat controlled the flame and did not cycle on and off. They might be able to give you ideas for the controls on your forge.

Lots of luck and keep us posted.
 
nc_cooter said:
Mike hit some good points. PID control of temperature is difficult at best. Temperature is slow, valves are fast.....relatively speaking. I worked for a number of years designing and building propane/air mixing systems to be used in place of natural gas. To achieve a constant mix at variable flows requires a specially designed dual body valve with a single valve stem. They were large systems capable of 200k scfm with a 20/1 turn down. I would say "it's possible but not practical.
Of course that was 25 years ago and technology has come a long way.

Mike Broach
Click to expand...

I agree with Mike on this. I also work in this industry and although possible, it would be relatively expensive to do. Not to say it can't be done, but may not prove to be practical.

Craig
 
C L Wilkins said:
I agree with Mike on this. I also work in this industry and although possible, it would be relatively expensive to do. Not to say it can't be done, but may not prove to be practical.

Craig
Click to expand...
C.L.

Who said anything about practical. I'm a knife maker. There is nothing practical about that.:D


Your point is taken though.

I have the forge body completed and will be getting started on the burner assembly this week.
At the very least I will have a digital read out hooked to the K thermocouple so I can adjust the temperature with some accuracy.
I am building this so I can do cans. My low boy doesn't have enough room.

A happy new year to you and yours, Fred
 
Here's a thought I have been toying with taken to the next level
Oxygen sensors for cars operate in a heated post combustion stream by generating small voltages from some sort of electrolytic reaction, I was thinking that I could set one up to monitor the atmosphere in my propane forge, after all what they are really looking for is perfect combustion (no leftover oxygen or fuel) to indicate a stochiometric mixture (I think 14 to one oxygen to fuel but I haven't been playing with hotrodding in a while) why not take it farther and use the O2 sensor to control an injector air butterfly. You could use off-the shelf cruise control components, a junkyard valve body, a junkyard o2 sensor (or a new one if you really want to spend the money, Summit racing sells O2 monitors that are really just a sensor and a 5 bar LED readout device to indicate rich , lean or good ((reducing, oxidizing or neutral)) I'm sure it cant be too hard to hook those up to a controller designed to hook up to a low voltage analog input((Type K thermocouple for instance)) ) use one controller to operate a needle valve for your gas to maintain temperature, the second for your atmosphere.

I was playing with gas forge building and atmosphere monitoring was one of the things I was goingto play with until my funds dried up.

If someone takes this idea and makes it work, please share with the forum so we can all benefit.

-Page
 
My suggestion on using a PId controller to maintain temperature on a forge is to use the controller to turn on/off the gas and the air at the same time. This way the forge pulses on to raise temperature and off when the temperature is reached, BUT THE GAS/AIR MIXTURE DOESN'T CHANGE. You will have to play with the hysteresis temps and times to get the temp range you want. THIS WILL ONLY WORK AS LONG AS THE TEMPERATURE OF OPERATION IS HIGH ENOUGH TO REIGNITE THE GAS/AIR MIXTURE!!!!!!!!!!!!! make sure you get the correct gas/air mix prior to using the PID.

Jim Arbuckle
 
The PID controlled system that I designed has the same parameters that Jim spoke of. The air is balanced by a variable power control (dimmer switch), and the gas feed is balanced by a needle valve.
Here is the improvement:
There are two parallel systems, one set to HI and one set to LOW. Each is pre-set to a proper ratio and atmosphere.The LOW setting allows the forge to run at a lower end temperature, say 1300F (unregulated), but keep a proper atmosphere. The parallel system has a solenoid valve on the gas supply and a second dimmer switch on the fan. When the PID relay switches on, the fan goes to high and the solenoid opens. The forge then runs at the rate this system is set at (lets say for purpose of discussion, that this setting would run the forge around 1600F if left running, but the atmosphere will still be balanced), and shuts the HI flame off when the PID setting is crossed (say 1475F). The flame them stays on at the LOW setting.Once the chamber temperature drops to the lower PID setting (say 1470F), it cycles back on to HI mode.

This all sounds complicated, but requires only one cheap PID ( and its K thermocouple), one cheap 1/4" gas solenoid valve, two 1/4"needle valves,one blower fan, and two fan controllers or dimmer switches ( two dimmers make it slightly more accurate,atmosphere wise, but one will work fine).The entire setup can often be built for less than $100.

This set-up allows the flame to never go out, preventing the WOOFFF, and loss of arm hair that often goes with it, upon re-ignition ( it isn't called detonation for nothing). It also cycles between a lower flame and a slightly ( how much higher depends on the gas/air adjustment) higher flame, thus avoiding the all or nothing heat input usually encountered in a solenoid controlled burner.The softer cycle makes for a closer and smoother heat control, and puts much less wear on the solenoid valve. With a little tweaking,and the blade in a pipe or muffle, 2F regulation in a gas forge is possible (5F is more practical and easier on the solenoid). Since the temperature control is based on the temperature of the thermocouple, placing the thermocouple in the muffle, where the blade is, lets you will know how hot the blade is, not how hot the flame is.

The other good thing is that this set-up can be built into a nice box or small cabinet, and moved from forge to forge (if you have multiple units). The air feed and gas feed hoses can be connected and disconnected with quick connects. The thermocouple can be hardwired,and moved from forge to forge, or disconnected with a thermocouple connector ( with a dedicated probe in each forge).
I did an overkill prototype in a briefcase. It had switches and lights, and two PIDs. Looked like,"The Man from Uncle", or "Mission Impossible".

Like I said, It sounds complicated, but is super simple. I'll look for a set of schematics and drawings I made a while back. They are somewhere around here.
Stacy

Note to Fred:
A venturi system requires a sufficient flow of gas to draw the venturi. If the gas is proportionately controlled by a motorized pressure regulator, as it cycled down it would reach a collapse point where the flame would back flash into the manifold (This is not good!!).Also the atmosphere would be all over the place. Try slowly turning the regulator down on a venturi forge to get it to stay at a low setting and you will see what I am saying. The only practical way to avoid this would be to use a blower.
 
This is great information. I new there had to be a fairly simple way to make this happen. I just could not put my finger on it.
Stacy, if you find those drawings I would be in your debt if you would pass them along to me.
I am a bit out of my element with the control part of this forge. But I am fortunate to have a good friend that is helping me with the electronics and this is his field of expertise.

Thanks again for all the input. It has made my day. :thumbup::cool: Fred
 
After several PMs for more information here goes:
In a normal blown burner set-up,the gas feed to the burner is adjusted with a needle valve to fine tune the gas flow (the pressure regulator keeps it to a maximum level).The blower is adjusted with a power controller (often just a dimmer switch). In a two mode set-up, there are two needle valves and two power controllers. One set is exactly the same as before. The added parallel system allows the two systems to be set at two levels, LOW and HI.All that is needed for the two to toggle from one to the other, is some way for the HI system to be turned on and off. By having the PID controller relay turn a solenoid gas valve on and energize the Hi power blower controller,this is easily accomplished. The low system is always on, the Hi turns on and off as needed.

If you want this to control the burner to keep an exact temperature, program the PID to turn the HI system on at 5F below the target, and off at 5F above the target. This will regulate the temperature +/- 5F. That is close enough for any HT specs we deal with.

If you want a system to switch your forge from a preset HI to idle(LOW), just set the PID for the HI desired, and add an On/OFF switch in the relay output line. Switching this switch to off will drop the system to LOW where it will idle until the switch is flipped back on,allowing the HI system to operate. This will save fuel when making damascus or forging. A foot operated switch (From HF,used to turn power tools on/off ) is the simplest way to control the system. If you are only switching from full blast to low, you actually don't need the high blower controller, but it is best to add it,though, so the fan can be best adjusted for a proper atmosphere.

To set up and operate the system, set the PID for 500F, this will keep the high part off. Adjust the LOW needle valve and LOW blower controller until the forge is running around the desired low point,say 1400F for example, and the atmosphere is balanced properly. Now raise the PID to 1600F, allowing the HI system to energize. Adjust the HI needle valve and HI blower until the burner is running balanced as it approaches 1600F. Drop the PID to 1500F, and the system will operate at 1500F with a continuously balanced atmosphere.With this set up you could run the PID at any temp from 1400 to 1600 and not need to readjust anything. Use the hi/low set up temps needed for the operating temperature range desired. Most forges can run from around 1000F to max temp with this system.Once you set up the burner balance, the only future adjustments will be to program the PID parameters, as long as you are in the general range of the original set up.

Attached is a simple schematic of the system (with parallel spelled wrong).

Any questions?
Stacy
 

Attachments

  • controller 001.jpg
    controller 001.jpg
    20.8 KB · Views: 458
Just a note, there are a couple types of dimmer switches (potentiometers). The one you want is not the light dimmer, buy one for a fan, I believe. I was just asking a guy at the electrical place about this a few weeks ago. He said the light dimmers work, but have a fairly short life if they are dropping too many watts. He said the fan? trimmers are better at staying alive for a decent length of time.
 
Good point, although for a small blower ,current draw is probably not a worry. I use the simple solid state speed control units available for about $5. An example is:
http://cgi.ebay.com/Solid-State-AC-...ryZ71393QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

Another way to make this set up work nicely ( especially for a small forge) is to find a two speed fan, like the ones in an exhaust hood.
While on the subject of fans, I often observe new makers wanting to use big blower fans. You don't need a big fan unless you are going to run a huge burner. Most blowers get choked down with a damper plate to a fraction of their actual capacity. 40-80 CFM is more than enough for most forges. 100-120 CFM will run a big burner rig.

One more note on why all this is important:
Many folks just make (or buy) a burner and hook it up to a blower. They crank the gas up until the forge looks like a jet taking off from the deck of a carrier. This may impress the neighbors kid, but does nothing for a knife blade ( I remember well the photo of one of our new makers proudly showing the 8 foot flame comming from the front of his new forge). A balanced flame has just enough extra fuel to have a small amount of flames outside the chamber. You don't need any extra air coming in from the back or front, the blower is supplying all the burner needs ( nor do you want to try and work in front of a hugh blow torch). Cutting the air and fuel back will allow the burner to run just hot enough to heat the forge chamber to the proper temperature, without excess oxygen (bad) or fuel (dragon's breath). The forge and burner will last a lot longer, steel will not be burning (decarb) so bad, and scale will be reduced. I read posts that say, " I built a forge last night and it can get up to welding heat in 3 minutes." Well, that is impressive, but what you really want to hear is, " I have been working on my forge for a few weeks now. It runs very quite, and after coming up to heat, holds at 1500F nicely. Even at welding heat,the outside barely gets warm."
Stacy
 
Bumped to answer more emails.
Seems like a lot of folks are interested in this.

There was some confusion caused by the way I quickly sketched the blower hookup. The TWO wires from the blower motor seem to be attached together. The common wire (marked as ground) is actually separate from the hot lead (the one hooked to the speed controllers).Sorry for the confusion. I'll try to make a larger and more detailed schematic later today.

Stacy
 
Fred,
If the controller cycles the flame size while you're in front of it, this would be a problem? Install an override to keep it in one state or the other?
I'd add mass to the end of the thermocouple to keep the cycling too fast.
 
howiesatwork said:
Fred,
If the controller cycles the flame size while you're in front of it, this would be a problem? Install an override to keep it in one state or the other?
I'd add mass to the end of the thermocouple to keep the cycling too fast.
Click to expand...

I believe it will take a bit of tweaking. I will be using this new forge for welding , almost exclusively. When I finally get it tuned to the heat range, I want to work in, I don't feel there will be that much of a pulse between the high and low settings. We'll see what happens. Of course, working with explosive gas, you don't want to leave too much to chance.
After reading bladesmiths post, on this, I think, since both the low and high settings are adjusted separately to an optimum burn, there should not be a great deal of dragons breath, to quote Stacy, to deal with.
I have a thermocouple housing welded into the forge body. It is a large commercial unit with a lot of mass. I believe you are correct; the cycling should be kept to as much of a minimum as possible.

Thanks for your input, Fred
 
You must log in or register to reply here.
Back
Top