making potassium (per) chlorate

[WSM]

Sadly, the second power supply had the same issue as the first; displayed an AC failure mode (my AC power is definitely within the supply's specifications) and no power output. I also contacted the manufacturer's service department and got some very good suggestions and guidance, but in the end, they also concluded it was a defective unit. The seller took the unit back and refunded my payment, so I gave them a good review.

Too bad about those supplies, I really could have used a power supply with that range and all those features.

I'm back to using the power supplies I have on hand (which is MANY, just not as versatile as the Sorensen would have been). If I get lucky and find a functional high end unit at a price I'm willing to pay, the scene will be different; but for now, I'll proceed with what I have on hand.

WSM

 

I've seen there are several different people/companies on eBay selling similar power supply units (Sorensen XFR 7.5-140) and I decided to go for it (again). The average prices range from $280 to $600, plus shipping.

 

One that caught my eye was asking $350 or best offer, including shipping. I had nothing to lose, so I offered $230 including shipping. They countered with $275 with shipping, which I figured is reasonable so I bought it. The shipping from their offices in Florida to my area for something weighing about 25 pounds is roughly $50 so I figure I got a very good deal. Another plus is that these sellers offer sales AND service (not just surplus like the last folks) so my chances were good to receive a functioning unit!

 

It arrived a couple days ago (in less than a week) and I added a power cord and tested the unit... it output power (well voltage anyway). I still need to apply it to a load (electrodes in a cell) and give it the "acid test" to prove it, but I am confident it's a good power supply.

 

The range is 0-7.5 Vdc and 0-140 Adc, and fully functional in the full range without the need to derate the output (according to the manufacturers representative). It also has CV and CC capability and is fully programmable; truly a top tier power supply. Plus, if I should want to expand the range of this supply, they are capable of being connected in either series or parallel to a second unit of the same type and range!

 

Besides being greatly versatile for all sorts of electrochemical experiments; If all goes well, I can use this supply to finally determine the capacity and heating rates of the various titanium tubing samples I have for use as leads for my electrodes!!!

 

WSM

Edited January 31, 2016 by WSM

[WSM]

I took some time yesterday and began to prepare the electrodes for my sodium chlorate experiment. My design is geared for higher output so I'm ramping up the size and capacity of the components. The power supply (mentioned previously) is a small regulated DC supply rated at 5 Vdc (adjustable between 4.2 V and 5.8 V) and a maximum output of 55 Amps.

 

Note: this is NOT the supply described January 30th.

 

I've sized my electrodes to demand 44 Amps so the power supply should run comfortably at 80% of it's capacity.

 

 

Electrode assembly materials

 

The electrodes will be spot welded to 1/2" (12.7mm) OD thin-wall CP titanium tubing, which will have the connection ends heated, press-formed flat and sealed (which takes multiple steps as described in Swede's blogs). After the electrodes are welded to the titanium tubing, the tubing will be filled with pure copper rods, 3/8" (9.525mm) diameter with the top 2" (~50mm) threaded with 3/8" SAE NC threads for connecting the input power (via a brass nut and two brass washers). The space between the OD of the copper rod and the ID of the titanium tube will be filled with lead free solder (95% tin) to stabilize the copper core and make good electrical contact.

 

The reason for the copper core is to increase the electrical conductivity of the electrode assembly and cut way down on the heating of the assembly under power. I've used the same type of setup successfully for the LD perchlorate electrodes and they did run cooler. The sealed titanium should fully protect the copper and tin core of the leads. I'm threading the top ends of the copper rods because the wall thickness of the titanium tube is 0.020" (0.508mm) and threading the ID of the titanium to accept a screw is not possible in this case.

 

The copper is hard so it needs to be annealed before cutting the threads (and possibly several times during the threading operation) and will make an excellent electrical connection when used as a filler for the titanium leads.

 

I'll post more as the assembly comes together...

 

WSM

Edited February 1, 2016 by WSM

[frederick]

A question, well I should say a math check, if someone with experience wouldn't mind.

 

I started a 4 gallon bucket cell 5 days ago. KCl saturation was done a 80F. After 5 days of running 5v and 17 amps I am just starting to see a little buildup of what I assume at this point is KClO3 in the cell.

 

By my math this is close to what I should see. The cell is running at about 100F and at this current input I should be turning around 163 grams per day. Solubility of KClO3 at this temp is 1995 grams (in 4 gallons of water) so I should just start to see precipitant, which I am.

 

Does all this sound correct?

[WSM]

A question, well I should say a math check, if someone with experience wouldn't mind.

I started a 4 gallon bucket cell 5 days ago. KCl saturation was done a 80F. After 5 days of running 5v and 17 amps I am just starting to see a little buildup of what I assume at this point is KClO3 in the cell.

By my math this is close to what I should see. The cell is running at about 100F and at this current input I should be turning around 163 grams per day. Solubility of KClO3 at this temp is 1995 grams (in 4 gallons of water) so I should just start to see precipitant, which I am.

Does all this sound correct?

Hi frederick,

 

On the face of it, it all sounds good. If you harvest the crystals, recharge the KCl levels in the cell and run it again, I'm sure you'll get more crystals sooner because of the precursor ions already there.

 

Try it and tell us what happens.

 

WSM

[frederick]

Thanks WSM.

 

I did harvest the cell over the weekend, gave it cleaning, recharged the KCl levels, and its back running. The KClO3 was cleaned, dried, and tested and I am happy with result, thus far.

 

Will see how round 2 goes after a few mods. I added some insulation to the cell to make up for the cold temps we have here and that increased the cell temp by about 30F, which also kicked the current up to 19, so a better result is expected.

[WSM]

Thanks WSM.

I did harvest the cell over the weekend, gave it cleaning, recharged the KCl levels, and its back running. The KClO3 was cleaned, dried, and tested and I am happy with result, thus far.

Will see how round 2 goes after a few mods. I added some insulation to the cell to make up for the cold temps we have here and that increased the cell temp by about 30F, which also kicked the current up to 19, so a better result is expected.

 

Excellent. Living in the South West US, I wasn't considering the weather in the more frigid regions. Yes, by all means insulate the cell. In the summer it shouldn't be necessary to isolate your cell from the ambient temperatures.

 

A fellow (from the mid west) at the PGI convention in Gillette, WY, set up his 1 gallon pickle jar cell in the manufacturing area to demonstrate it (he was selling kits in the flea market, also). I noticed he used aluminized Mylar bubble wrap insulation material (available at big box home centers), and I bought a small roll of the same material when I got back home. As long as the mother liquor doesn't touch it, the aluminum part won't be negatively affected (aluminum breaks down in an alkaline environment).

 

Keep up with your cell and share how subsequent runs go as you fine-tune the system.

 

WSM

[WSM]

I managed to score some platinized CP titanium mesh anode material for a reasonable cost. The stock is sized to demand about 20 Amps at 0.2A/cm², and is coated over the entire surface. The nice thing is the stock doesn't have a worthless strap attached so I won't have something to cut off before I can spot weld a tubular titanium lead to it to finish the anode.

 

I'll post more about these later as I resume my perchlorate research.

 

WSM

[frederick]

Hi frederick,

On the face of it, it all sounds good. If you harvest the crystals, recharge the KCl levels in the cell and run it again, I'm sure you'll get more crystals sooner because of the precursor ions already there.

Try it and tell us what happens.

WSM

Everything turned out well, better than expected actually. Seven days run at 19 amps produced about 1400 grams.

[WSM]

Everything turned out well, better than expected actually. Seven days run at 19 amps produced about 1400 grams.

 

That looks great. Are you going to recharge the electrolyte and run it again? If you add in a pH control scheme, I expect you'll see an even greater yield and lower power consumption.

 

Keep up the good work.

 

WSM

[frederick]

Yes it will be recharged and rerun. I have two buckets in rotation, one is in service while the other is recharging at room temp, and will continue the 7 day rotation.

 

In this config I can pull the running cell and harvest the KClO3, insert the recharged cell, do a second harvest the next morning of room temp KClO3, then recharge it at room temp over the 7 days it is resting which is helpful considering the chunky KCl used.

 

If the yield is consistent I won't incorporate ph control, in this configuration I can just set it and forget it, and if my demand exceeds the supply I will simply put another cell into operation. The cost is minimal at this return and its no muss no fuss - which I like.

[WSM]

Yes it will be recharged and rerun. I have two buckets in rotation, one is in service while the other is recharging at room temp, and will continue the 7 day rotation.

In this config I can pull the running cell and harvest the KClO3, insert the recharged cell, do a second harvest the next morning of room temp KClO3, then recharge it at room temp over the 7 days it is resting which is helpful considering the chunky KCl used.

If the yield is consistent I won't incorporate ph control, in this configuration I can just set it and forget it, and if my demand exceeds the supply I will simply put another cell into operation. The cost is minimal at this return and its no muss no fuss - which I like.

 

I calculated your cell yield (if run continuously) would be about 160 pounds per year. There are ways to modify and optimize the system, but if you're happy with it, keep it the way it is.

 

Edit: Keep us up to date with any changes in performance and modifications that may need to be made. If things just keep going as they are now, let us know that too. It's good to keep up on the progress of fellow electrochemists.

 

WSM

Edited February 27, 2016 by WSM

[WSM]

I took some time yesterday and began to prepare the electrodes for my sodium chlorate experiment. My design is geared for higher output so I'm ramping up the size and capacity of the components. The power supply (mentioned previously) is a small regulated DC supply rated at 5 Vdc (adjustable between 4.2 V and 5.8 V) and a maximum output of 55 Amps.

5 Vdc 55 Amp Supply.JPG Note: this is NOT the supply described January 30th.

I've sized my electrodes to demand 44 Amps so the power supply should run comfortably at 80% of it's capacity.

Electrode Parts Preparation.JPG

Electrode assembly materials

The electrodes will be spot welded to 1/2" (12.7mm) OD thin-wall CP titanium tubing, which will have the connection ends heated, press-formed flat and sealed (which takes multiple steps as described in Swede's blogs). After the electrodes are welded to the titanium tubing, the tubing will be filled with pure copper rods, 3/8" (9.525mm) diameter with the top 2" (~50mm) threaded with 3/8" SAE NC threads for connecting the input power (via a brass nut and two brass washers). The space between the OD of the copper rod and the ID of the titanium tube will be filled with lead free solder (95% tin) to stabilize the copper core and make good electrical contact.

The reason for the copper core is to increase the electrical conductivity of the electrode assembly and cut way down on the heating of the assembly under power. I've used the same type of setup successfully for the LD perchlorate electrodes and they did run cooler. The sealed titanium should fully protect the copper and tin core of the leads. I'm threading the top ends of the copper rods because the wall thickness of the titanium tube is 0.020" (0.508mm) and threading the ID of the titanium to accept a screw is not possible in this case.

The copper is hard so it needs to be annealed before cutting the threads (and possibly several times during the threading operation) and will make an excellent electrical connection when used as a filler for the titanium leads.

I'll post more as the assembly comes together...

WSM

 

I've started getting my electrodes put together and next I'll populate the lid of my sodium chlorate cell (a 24" long piece of schedule 40, 12" PVC pipe with 1/2" thick PVC plate for a base) with all the fittings for a vent, the electrodes and various other sensors, sampling ports, etc.

 

I'm taking a few photographs to add to the article I'm writing for the PGI Bulletin (Homegrown Oxidizers, Part 12) and will post a few here as well.

 

Between physical ailments, work and travel, my progress has been slower than normal, but I'm grateful for the free time I do get where I can make progress on my research and writing. I'll add more as I get further along...

 

WSM

Edited February 27, 2016 by WSM

[WSM]

I made some progress yesterday, but not as much as I'd hoped for. I managed to cut and form several pieces for the anode and boxed cathode for the 44 Amp electrodes to be used in the experimental sodium chlorate cell.

 

I started by facing the ends of the thin wall titanium tube on my lathe. Next I set up a one ton arbor press with a section of cold-rolled steel to form the closed end of the CP titanium tubing.

 

 

Next I heated the end of the tubing with a propane torch to orange hot.

 

 

While the tube is hot I press about 5/16" (~8mm) of the tube end till it resists, reheat and press again alternately till the end is pressed flat.

 

Edit: if pressed too far without heating, the titanium tubing tends to split open rather than keep forming. This ruins the tube for use in a cell unless the split can be sealed with TIG welding. With alternate heating and pressing steps, the forming can be accomplished without the titanium fatiguing and splitting (see Swede's Blog).

 

 

After pressing, I buff the scale off the flattened ends and clean up the entire surfaces of the two titanium lead tubes.

 

 

 

The rest of the time was involved in a lengthy process of cutting several side ears and cutting and forming matching "C" shaped boxing spanners for attaching the cathode plates together where they "box" the anode (surround the anode electrically). I use Bi-metal hacksaw blades in a hacksaw frame to cut CP titanium. It cuts about like mild steel.

 

 

I didn't finish assembling the electrode assemblies because I couldn't reach my 3/8" tap to thread the ends of the copper fill rods. After they're threaded, I'm considering tinning the copper for better electrical contact with the insides of the two titanium tubes and the fill solder. Well, actually I feel the solder fill will go smoother if the copper rods are tinned, plus this step will help avoid gaps in the solder fill.

 

After the electrodes are completed, I still need to drill and tap holes in the 1/2" thick PVC lid for the cell for the myriad fittings and adapters to be placed there.

 

In the blogs I mention using special drill bits for safely drilling the PVC without damage. One type of drill I neglected to mention are simple spade bits. If drilling for standard plumbing sizes, I recommend getting a set of high quality spade bits with as large a number of sizes as you can. I find several sets omit some very useful sizes such as 7/16" (for 1/4" NPT tapping) as an example. Sets with steps of 1/16" between the bit sizes are the most useful, and the spade bits do a fair job of drilling hard PVC without damage to the surrounding material.

 

There's much more work to do before I can run the cell and test my ability to make my own sodium chlorate feed stock, but I'm making progress.

 

More to come...

 

WSM

Edited February 28, 2016 by WSM

[MrB]

I didn't finish assembling the electrode assemblies because I couldn't reach my 3/8" tap to thread the ends of the copper fill rods. After they're threaded, I'm considering tinning the copper for better electrical contact with the insides of the two titanium tubes and the fill solder. Well, actually I feel the solder fill will go smoother if the copper rods are tinned, plus this step will help avoid gaps in the solder fill.

 

Perhaps not feasible, the tin might run out from the end to fast, but, cant you tin the copper, and then melt tin in the Ti tube, with the copper rod above? Once the tin melts, the rod "should" slide in, the tin raise uniformly, and eliminate any air-gaps completely.If you can use flux on the inside of the tube, and put some on the top of the tin, that should help it flow nicely as well.

And, if it "fails", the copper rod doesn't want to slide in, or the tin flows out to fast on the crimped shut end, it's just a matter of reheating, and starting over, so to speak.

 

B!

[WSM]

Perhaps not feasible, the tin might run out from the end to fast, but, cant you tin the copper, and then melt tin in the Ti tube, with the copper rod above? Once the tin melts, the rod "should" slide in, the tin raise uniformly, and eliminate any air-gaps completely.If you can use flux on the inside of the tube, and put some on the top of the tin, that should help it flow nicely as well.

And, if it "fails", the copper rod doesn't want to slide in, or the tin flows out to fast on the crimped shut end, it's just a matter of reheating, and starting over, so to speak.

B!

 

I should mention that I press the end completely flat, so the solder won't leak out (and the cell liquids can't seep in). If the end isn't air tight when I'm done flattening it, my next step is to spot weld all across the edge so it is air tight. Next comes spot welding the electrodes to those sealed edges.

 

Tinning the copper allows for smoother addition of the solder to the titanium tube, without bubbles on the copper surface, which would effectively lower the conductivity of the assembly. The solder doesn't wet the titanium but is only in physical contact with it. The copper, on the other hand, is in intimate contact with the solder fill, and without any gaps will optimize the conductivity of the whole assembly

 

When I insert the copper filler, is after putting in about an inch or two of molten solder. While heating the titanium tube with a propane torch, the tinned copper rod is carefully inserted till it reaches the bottom of the titanium tube, where it contacts the sealed end inside. The heat is slowly run from the sealed portion of the titanium tube, with circular motions till the solder fills the entire titanium tube, adding extra solder as needed to the top of the tube, ensuring the maximum electrical contact within the whole electrode assembly.

 

I've done this several times before and have the technique down; and it definitely reduces the resistance of the titanium (thereby reducing heating as the current flows).

 

WSM

Edited March 2, 2016 by WSM

[WSM]

I expect the electrodes will be completed soon so I can finish the assembly of my cell and put it into service. I have to learn the details of sodium chlorate making and purification so I can get back to my perchlorate experiments (I've exhausted my meager supply of sodium chlorate and need an affordable source to continue my work).

 

WSM

Edited March 1, 2016 by WSM

[MrB]

Sounds like you got a working strategy you like. Go with it :- )

B!

[WSM]

Sounds like you got a working strategy you like. Go with it :- )

B!

 

I've given this process a lot of thought for a long time, so I have a pretty good idea of what I'd like to do. All the minute particulars haven't been ironed out yet, so ideas and contributions of different methods are always helpful.

 

As I've progressed along this path, ideas from others that I've initially dismissed, have all too often been adopted by me later as the better method or approach. Thankfully, I'm not too proud to admit when I'm wrong; and in gratitude, I always like to give credit where it's due.

 

Thank you all for your help and helpful suggestions. Even if I don't always appreciate it when it's given, sooner or later I usually come around and follow the better route to succeed in what we're trying to do.

 

Please keep the ideas coming..., among them are likely to be a gem that opens the door to success.

 

Regards,

 

WSM

[WSM]

Perhaps not feasible, the tin might run out from the end to fast, but, cant you tin the copper, and then melt tin in the Ti tube, with the copper rod above? Once the tin melts, the rod "should" slide in, the tin raise uniformly, and eliminate any air-gaps completely.If you can use flux on the inside of the tube, and put some on the top of the tin, that should help it flow nicely as well.

And, if it "fails", the copper rod doesn't want to slide in, or the tin flows out to fast on the crimped shut end, it's just a matter of reheating, and starting over, so to speak.

B!

 

Thanks for the suggestions MrB.

 

It appears that my verbal description of the process I use wasn't clear enough to accurately paint a view of what I'm doing. It would be easier to show the steps in a video, if I were set up to do such a thing. For now, it's all I can do to just get things together and make them run.

 

Maybe someone out there will do a step by step demonstration of how these things are done, so it'll be easier to follow.

 

 

...Anyone...?

 

 

WSM

Edited March 2, 2016 by WSM

[WSM]

I just sent Homegrown Oxidizers Part Twelve off for final review before submission to the PGI Bulletin. I'm making serious progress on getting the cell ready to run and hope to have it operating before Easter (we'll see..).

 

Every time I make up a set of tubular leads, I seem to forget exactly how much work is involved. Man, those things are labor intensive (and costly). Still it's a pleasure to have them sealed in the cell lid and gas tight, so no salt creep or power lead corrosion.

 

I'll soon post more about my progress and some photos, too.

 

WSM

Edited March 6, 2016 by WSM

[Arthur]

Have you ever used plain Ti rod as the electrode hangers? Cost of work vs cost of metal.

[WSM]

Have you ever used plain Ti rod as the electrode hangers? Cost of work vs cost of metal.

 

Yes, early on.

 

In my thinking, the major consideration is conductivity. Copper is more than 32 times as conductive as titanium (Cu = 100, Ti = 3.1), and much less prone to heating. In low voltage / high current applications, this can be a very important factor; especially in the larger sized systems.

 

Maybe it's just me, thinking like an electrician. Also, I like to take a broader view and consider the tank construction and fitting materials and their thermal limitations.

 

Perhaps, I should set up an experiment and try both ways and see if my concerns are well founded or whether equally sized solid titanium leads would perform better than I imagine? It would be a lot less work using solid titanium, round rod leads.

 

One day maybe, but for now I'm invested in the system I've set up.

 

WSM

Edited March 6, 2016 by WSM

[WSM]

I made some progress yesterday, but not as much as I'd hoped for. I managed to cut and form several pieces for the anode and boxed cathode for the 44 Amp electrodes to be used in the experimental sodium chlorate cell.

I started by facing the ends of the thin wall titanium tube on my lathe. Next I set up a one ton arbor press with a section of cold-rolled steel to form the closed end of the CP titanium tubing.

I Arbor Press.JPG

Next I heated the end of the tubing with a propane torch to orange hot.

IMG_1229.JPG

While the tube is hot I press about 5/16" (~8mm) of the tube end till it resists, reheat and press again alternately till the end is pressed flat.

Edit: if pressed too far without heating, the titanium tubing tends to split open rather than keep forming. This ruins the tube for use in a cell unless the split can be sealed with TIG welding. With alternate heating and pressing steps, the forming can be accomplished without the titanium fatiguing and splitting (see Swede's Blog).

IMG_1230.JPG IMG_1231.JPG

After pressing, I buff the scale off the flattened ends and clean up the entire surfaces of the two titanium lead tubes.

IMGP7828.JPG

The rest of the time was involved in a lengthy process of cutting several side ears and cutting and forming matching "C" shaped boxing spanners for attaching the cathode plates together where they "box" the anode (surround the anode electrically). I use Bi-metal hacksaw blades in a hacksaw frame to cut CP titanium. It cuts about like mild steel.

IMGP7829.JPG

I didn't finish assembling the electrode assemblies because I couldn't reach my 3/8" tap to thread the ends of the copper fill rods. After they're threaded, I'm considering tinning the copper for better electrical contact with the insides of the two titanium tubes and the fill solder. Well, actually I feel the solder fill will go smoother if the copper rods are tinned, plus this step will help avoid gaps in the solder fill.

After the electrodes are completed, I still need to drill and tap holes in the 1/2" thick PVC lid for the cell for the myriad fittings and adapters to be placed there.

In the blogs I mention using special drill bits for safely drilling the PVC without damage. One type of drill I neglected to mention are simple spade bits. If drilling for standard plumbing sizes, I recommend getting a set of high quality spade bits with as large a number of sizes as you can. I find several sets omit some very useful sizes such as 7/16" (for 1/4" NPT tapping) as an example. Sets with steps of 1/16" between the bit sizes are the most useful, and the spade bits do a fair job of drilling hard PVC without damage to the surrounding material.

There's much more work to do before I can run the cell and test my ability to make my own sodium chlorate feed stock, but I'm making progress.

More to come...

WSM

 

Yesterday, I continued the work on these electrodes and it came together like this:

 

I found that the thin walled titanium tube made a thin flat end to connect to the electrode. The problem is the material heats too much and too fast when spot welding. My answer to the problem was to back the material with similar stock; MMO ribbon in the case of the anode, and extra CP titanium sheet metal in the case of the cathodes.

 

 

 

These two photos show the lead tube end sandwiched between the MMO mesh and the MMO ribbon.

 

 

This photo shows the lead tube end sandwiched between the cathode and a piece of the same CP titanium stock.

 

I noticed that the thin walled titanium tube was harder to flatten to the point of being completely sealed than thicker wall titanium tubing. When I heat and flatten the tubing, I usually quench the heated end in water and test the closure by blowing through the tube while the end is submerged. If I see bubbles, I reheat, carefully tap the end with a ball peen hammer on an anvil and test again.

 

With thick walled tubing this completely seals the flattened end, but with the thin walled tubing, no. The fix is to spot weld a row of welds across the flattened end of the tube and Viola, it's sealed shut (air tight).

 

 

The rest of the photos show the cathode "box" assembly and the finished electrodes together (the anode is wrapped with a paper towel to prevent scratching the MMO surface on the cathodes).

 

 

 

 

I still need to tin the copper fill rods and fill the titanium tubes. I was waiting to thread the ends of the copper rods but decided to use copper electrical clamps with a spade connection on them which will simplify the power connection to the electrodes without extra work.

 

 

There's still more to come...

 

WSM

Edited March 6, 2016 by WSM

[WSM]

Sounds like you got a working strategy you like. Go with it :- )

B!

 

Thanks. I know the way I'm doing things is more involved than what most folks want, or are willing to do, but I show them as what one man is doing.

 

Perhaps it will inspire others, or maybe just help them overcome some small technical challenge they face with their system. Either way, I hope it helps.

 

Most of what I've developed on my own is through overcoming technical difficulties while building systems for my own research. If by sharing, it helps others, I'm glad I could help. We'll all learn together, even if it's "I don't want or care to bother with such a complicated approach". At least the information is out there...

 

WSM

[WSM]

Homegrown Oxidizers Part Twelve has been submitted to the PGI for addition in the next published Bulletin. The next article will show the cell being completed and put into service, making sodium chlorate.

 

I have plenty to learn about making sodium chlorate and purifying it for use in the perchlorate cells. Folks in the US seldom, if ever, bother with sodium chlorate, and usually make potassium chlorate directly from potassium chloride. This is virgin territory for me, but I hope the learning curve is short.

 

We'll see...

 

WSM

Edited March 7, 2016 by WSM