Category Archives: Projects

Hacking / Modding the Bissell Little Green Cleaner

 

I bought my first Little Green Machine in 1994.  It was a first-generation, ugly, dark green box with one purpose: to suck up pet accidents and spills.  It was the only thing on the market at that time, and I desperately needed something to assist in cleaning up after an aging dog with bladder control issues.  It worked well, and I got several years’ use out of it for my $70.00 investment.  Eventually a plastic hose barb cracked inside the unit and I couldn’t find a replacement so I trashed it.  Bissell has always had a policy of refusing to supply internal repair parts to customers, thanks to paranoia about consumer lawsuits.

I purchased a newly redesigned model at that time—one that sort-of resembled an egg standing on its fat end.  It was a poor redesign with some flaws (like solution heaters that constantly failed) that I made do with for several years.  I must say here that Bissell was good about support.  After writing to complain twice about product failures, they replaced the unit both times at no charge.  The second exchange was for the first version of their 1400-series model (Figure 1) , which at the time was brand-new.

The Bissell Little Green

Figure 1. The Bissell Little Green.

I’ve been through three of these in the last six years.  After the first unit of the new series died, I began studying their design to see how I could service it myself (and extend its life a few years as well).  These things have gotten progressively more expensive, and my last replacement was priced just over $100.00.  For that amount of money I expect something to last longer than a year!

Quirks and Design Flaws

The first redesigned 1400-series model had a simple storage clip on its backside that you’d snap the hose’s sprayer wand into.  The wand was always popping out of the clip and hitting the floor, which would eventually crack the sprayer head and further decrease its already weak suction.  Other problems included: the solution heater would die, the sprayer would simply stop working, or would begin to clog after using the unit for six months or so.  The wand would drip cleaning solution after you finished using the unit and were putting the wand away.   Finally, the cleaner would develop a stench from the urine I’d be cleaning despite rinsing the unit‘s recovery tank after each use.  This is due to the way the recovery tank is designed.

Phew!

Figure 2. Tank areas that collect waste material.

You can see by looking at Figure 2 that there are a couple of problem areas (circled in red) near the recovery tank’s top.  These seem to collect material that eventually begins to smell.  Simply swishing hot water around in the tank is not sufficient to clean the areas, and you cannot access the tank’s insides to physically wipe the surfaces clean. The sprayer head and wand connection point was designed such that dirty water tends to collect in the nooks and crannies which also creates an unsightly, stinky mess.  This junction also provides a break in the air stream and contributes to the unit’s biggest problem– it simply doesn’t have much suction power.  The area you clean will remain fairly damp.

Design Improvements in newer models

I kept that original 1400-series model going for three years, by my estimates about two times longer than I should’ve been able to.  The fixes were fairly simple for a geek to do.  The recovery tank sat on a very thin rubber gasket that was glued to the main unit’s base.  This was in addition to the rubber gasket on the tank itself.  The base unit’s gasket began to wear away after a time, and I managed to make a replacement from some sheet rubber I purchased off of eBay.  I cracked the sprayer head attachment apart and re-glued the pieces together with five-minute epoxy.  I also glopped extra epoxy into the designed-in gaps and cracks while I was at it, which helped increase suction.  I finally had to retire the unit after trying to discover why its motor was whining so loudly.  After I took the unit apart to possibly oil the motor I learned more about the unit’s Achilles’ heel.

The currently-sold unit has several design revisions that seem to address some of the problems.  The wand clip on the unit’s back has been redesigned and includes a swivel arm that locks the wand in place.  Hallelujah!  No more dropped sprayer wands!  The sprayer hose has been changed from a solid green to a greenish-tinted clear hose.  This allows the user to see how much crap is in the hose (and tell how well the suction is working).  The recovery tank’s base gasket is gone (shown in Figure 3 sans gasket), which also eliminated a source of fluid leakage into the base.

No more gasket!

Figure 3. Redesigned recovery tank dock on base.

Finally, Bissell redesigned the sprayer head with better welds and eliminated the extra gaps.  This helps to a degree, but the redesign introduced several new ways for dirty solution to collect in the sprayer/wand connection and stink up the works.  They did nothing to redesign the collection tank tower, which still collects material that leads to ‘Stinky Tank syndrome.’  My current unit had its hose replaced twice, once under warranty.  (This may have been due to my sucking up hot water in order to flush and de-stink the unit.)  The original hose was a little too short and it was under stress when the wand was placed into the clip.  The stress also ended up breaking the clip after the unit was no longer under warranty.  I performed a couple of simple fixes to increase its suction but didn’t implement anything major.  I performed the second hose replacement myself, as I finally found a source for internal parts.  I had to buy a second unit while waiting for the parts to arrive as they were backordered for a month-and-a-half.

 

Why doesn’t it suck (enough)?

The Bissell Little Green’s designers had a challenge, I’m sure…how could they suck up a spill and easily separate the water from the air?  Most wet/dry vacs have a foam filter that the suctioned air passes through, which keeps the water out of the exhaust.  I’m sure the designers didn’t want their customers to deal with replacing a filter (liability lawsuits, you know!), given the tendency for most consumers to screw up the simplest of operations.  This probably led to the current design.

Suction is provided by a motor with a squirrel cage fan attachment located in the base.  Air and water are pulled into the sprayer head, through the hose and into the recovery tank.  The tower-like structure you saw in Figure 2 is divided in half.  The water falls into the tank since it’s heavier than the air, and the air exits the hole in the opposite side of the tower.  The air stream then passes the fan/motor assembly and is blown out the unit’s backside. Simple design, yes, but flawed.  The air has a high degree of moisture in it, and there is no filter of any kind to protect the metal motor and blade assembly.  If you have too much suction the air will have even more water in it; this will accelerate the disintegration of the motor and allow water to blow out of the exhaust port with the air.  You read that right—the motor will disintegrate faster.  This is the Achilles’ heel I mentioned earlier.  When I disassembled my last unit to try and oil the motor, I discovered that the blades and motor casing were so rusty that they practically fell apart in my hands.  That was what caused the motor to scream and whine.  This is a marketing textbook example of built-in obsolescence!  At the time I didn’t know where I could get repair parts, but now I know and will share the source with you.

Making Things Suck More

So, we know that the unit will eventually stop working due to the disintegration of the motor.  Should that stop us from making the unit work a bit better? NO!  If it’s going to have a finite life before we have to repair it, let’s mod it to perform its job better.

O ring to replace is seen vertically in middle of picture.

Figure 4. The Sprayer/Wand junction.

Look at the sprayer/wand connection in Figure 4.  You can see a rubber O ring that attempts to seal the wand/sprayer connection point.  In my last unit this O ring was so loose that no seal was taking place.  I went to the hardware store with the old ring and found a new one that was a little thicker.  Its size is given as a “number 32.”  I replaced this ring, which now requires the user to jockey around a bit when removing/reattaching the sprayer head.  The improvement is well worth the additional effort that will be required.

Now take a close look at the sprayer head/wand.  If you’ll notice, there’s a button that you have to press down on to remove the sprayer.  This is located on a molded plastic lever that flexes when the button is depressed.  This design feature also allows a good deal of suction to exit the hole around the lever.  Additionally, the end of the wand does not seal inside the sprayer head.  This is a second point that allows air to exit around the lever and allows dirty water to accumulate.  Sealing that leak will require a bit more work; however, we can significantly increase the suction by installing a flat rubber seal into the wand below the button/lever assembly.  How much additional suction does that give?  I don’t have a scientific method to gauge it, but by simply holding the sprayer head against my hand with the unit running before and after the mod, it feels as if I’ve approximately doubled the suction.

I took a piece of 1/16” thick sheet rubber and cut a 1.5-inch by 3.5-inch piece from it.  I then applied rubber cement around the edges of the piece (Figure 5) leaving the center dry.

Put glue only on edges of pad, nNOT in center.

Figure 5-a. Apply cement to pad edges on one side only.

 

See? Nothing in the center...

Figure 5-b. The finished pad.

This is important!  If the rubber adheres to the lever, it will be extremely difficult to work the lever and remove the sprayer head.  Next, I folded a piece of paper from a magazine (Figure 6) and inserted it into the slot around the lever.

Protect the lever from glue!

Figure 6. Protecting the lever from glue application.

This allows you to apply contact cement around the lever without accidentally getting glue on the lever itself.  Apply rubber cement to the inside of the wand in approximately the same area where the rubber will go as in Figure 7.

Carefully put glue into the wand...

Figure 7. Apply glue into the wand.

After applying the cement into the wand, quickly pull the paper out of the slot.  Now wait about five minutes to allow the glue to get tacky.  After you’ve waited, flex the rubber piece in an arc as shown in Figure 8 and carefully insert it into the end of the wand with the glue side facing the glue inside the wand.  DO NOT ALLOW THE PIECE TO CONTACT THE GLUE INSIDE THE WAND UNTIL IT’S IN POSITION.  If you do, it will be extremely difficult to remove.  You may want to use forceps or tweezers to assist you in alignment.

Be careful! Pad can stick where you don't want it to go!

Figure 8. Carefully placing the pad into the wand.

Once it’s aligned, press the rubber into place, exerting pressure around the edges of the rubber.  You may have to use a tool to reach in and push the edges down.

Made it!

Figure 9. The pad in it's proper place.

Let the piece sit overnight to dry; then reattach the sprayer head and turn the unit on.  You can hear a little bit of air still exiting the hole around the button.  Hold the sprayer against one hand and cover that hole with your opposite thumb.  If you tried the suction before the mod, you should notice a considerable difference in the amount of suction after the mod.  If you wish to cover the hole with a piece of electrical tape you can easily seal the leak, and remove the tape when you wish to clean the unit.  Alternatively, simply cover the hole with your thumb as I do while using the Little Green.  I am considering milling a second channel around the end of the wand with my Dremel tool, then using a second O-ring to completely seal the connection.  This would allow for more comfortable use of the sprayer wand.

Some tips on using the Green Machine

1)  Clean the unit after each use by holding the sprayer head under warm (not hot! I learned my lesson…) running water.  I will sometimes mix some pine-scented cleaner with water in a bucket and pull that through the hose. Note that you can only suck so much fluid into the recovery tank; there’s a white line on the tank showing the maximum level.  When you’ve finished flushing the sprayer head and hose in this manner, hold the sprayer head above the unit pointing upward for a moment or two to allow all the remaining water in the hose to be sucked into the recovery tank.

2)  Stopping the dribbles. When you’ve finished cleaning a spot and you’ve turned the power off, hold the sprayer head above the unit pointing upward and depress the trigger. Hold the trigger in for a count of five before releasing it.  MAKE SURE THE POWER IS OFF BEFORE DOING THIS OR YOU’LL SPRAY HOT CLEANING SOLUTION ON YOURSELF OR SOMEONE ELSE. Performing this maneuver allows any cleaning fluid still in the hose/sprayer head to drain back into the unit, and prevents leakage when you’re moving the sprayer head around to clip it onto the unit.

3) Clean the recovery tank after each use. Do not allow material to stay in the recovery tank!  This allows the material to soil the plastic and contributes to ‘Stinky Tank syndrome.’  Dump the recovered material into the toilet and flush it. Use the toilet because any pet hair in the water may clog your sink drain.  Run hot water into the tank from the top hole, then turn and twist the tank to thoroughly swish the water around before emptying the tank again into the toilet.  Remove the rubber gasket from the bottom of the tank and clean it with soap and water or pine-scented cleaner as well.  Material tends to collect on the backside of that gasket and also creates a smell.  If you have a bucket that’s big enough, you can dump the entire tank into a water/pine cleaner mix and get things smelling a bit better.

4)  Use only distilled water in the unit. I cannot stress this enough.  The minerals in tap water can build up after awhile and collect in the heater assembly.  This can clog the sprayer when bits of the accumulated minerals fracture off and mix with the cleaning solution. When the sprayer wand gets clogged you’ll see that fluid has entered the head but nothing exits when you press the trigger.   To remove the clog you can usually insert a cotton swab into the channel inside the sprayer head, and gently move it around to catch and remove the clog.  Occasionally the clog gets wedged into the sprayer hole; this requires pushing a slightly-bent straight pin into the hole from the outside to push the clog out. You can then use the cotton swab to remove it.

The heater itself can be cleaned by first disassembling the unit and then removing the cover on the heater but it’s a major hassle.  The heater box sits atop some plumbing bits and has a thin hose entering and exiting on its sides. To get inside it you remove the screws that hold the top cover on.  You’ll see maze-like channels inside the heater, and should see whitish material inside those channels.  Those are the mineral deposits, and you remove them by scraping them out.  You’ll then replace the top cover and reassemble the base.  This can take up to an hour the first time you do it.

 

Getting Repair Parts

I don’t know why it took so long for me to find a parts dealer (despite Google), but Hesco Sales (www.hescoinc.com) will sell you any part they can get from Bissell. There’s a section on their website where you enter your unit’s model number and can pull up exploded diagrams of the unit to help identify your part. In my case the part data wasn’t available on their site. Several emails later they had the part information and made the part orderable. Note that they have a $25.00 minimum order, so my advice is to get a couple of the hose clips I mentioned earlier. They’re inexpensive and make good filler material for your order. Besides, you’ll eventually need them.

Hopefully you found this information useful, and maybe you will be able to save your unit from the trash heap (or have gotten the courage to rescue one from the thrift store).  If one person can keep their unit running longer I’ve done my job.

Advertisements

18 Comments

Filed under Projects

An X10 Tip—Delaying a Trigger Signal to a Powerflash Module

I recently upgraded my home theater to bring it current with high-def content.  My 12 year-old 50” standard-def rear-projection set was replaced by a 60” plasma display, and my non-HDMI capable A/V receiver was dumped in favor of an Onkyo 7.1-channel preamp/processor.  (Naturally, the 9.2-channel version with Internet streaming audio came out a month later.)  This gave me an opportunity to finally install the separate power amps I purchased as surplus from my employer several years ago.

After opening the pre/pro and perusing the manual I realized I’d need a method of controlling the amps.  The Crest Audio Vs-450s have mechanical power switches and no trigger inputs.  The Onkyo has three trigger voltage outputs that are programmable as to when they turn on.  I needed some sort of interface between the pre/pro in my living room and the amps, which I located in a basement equipment rack.  Basement installation was necessary because the amps have big cooling fans in their front panels.  I went to the web and shopped around for trigger voltage-controlled power strips.  Several companies make them, and they’re pricey (around $200 for a two-outlet unit).  That was out of the question.

I then considered building my own trigger-controlled outlets.  I found several schematics and blog posts from people who’d built their own, but I really didn’t want to tackle yet another project.  Then inspiration hit me—I use X10 for home control, so why not use it to control the amps?  I ordered three Powerflash modules and five X10 appliance switches, and sketched out how everything would connect.  The X10 Powerflash modules have two screw terminals for an input, and can take either a control voltage or a simple switch closure (selectable by a switch on the module).  The voltage input would come from an Onkyo trigger output.  The module has three possible modes of operation; one mode will send an X10 ‘on’ command when voltage is first applied and an ‘off’ command when the voltage is dropped.  Each amp would be plugged into an appliance module and be set to a unique address that matched one of the Powerflash interfaces.  I could theoretically turn amps on and off when needed depending on the listening mode.  For example, if I wished to listen to 2-channel music I could turn on only the front left – front right amp.  When I switch to a movie surround mode other amps would come on.  This would allow a degree of ‘green’ operation and save some money on the power bill.  It was the perfect solution (or so I thought)!

I wired everything together and turned the system on for the first time, but the amps didn’t power up.  I could turn them on with a wired mini-controller but the Powerflash modules didn’t seem to work.  After a bit I had a forehead-slapping moment.  My DirecTV DVR is connected to a UPS in the equipment rack, and UPSes are notorious X10 signal eaters.  I dug out a filter specifically designed for signal-interfering devices and plugged the UPS into it.  No change—the Powerflash modules still didn’t work.  Then I realized that the APC  home theater power bar I’d installed would also filter out X10 signals.  To get around this I ran wires from the Onkyo trigger outputs through the wall and down into the basement.  I mounted an unfiltered power strip on a floor joist below the living room and plugged the Powerflash modules into it.  The power strip was then plugged into a basement AC outlet on a separate breaker from the entertainment system (but on the same phase in the breaker box).  Success!  The amps would then turn on with the pre/pro.

I spent some time listening to the new system and was very pleased with what I heard.  Even without calibration it sounded fantastic.  I was ecstatic!  Then I turned the system off and went downstairs for something, and that’s when I discovered that the amps hadn’t turned off.  After several hours of testing I again found that the amps would turn off with the mini-controller in the living room, so I used that as a stopgap measure while I thought the problem through.  Several days later I realized what the problem was.  The amps would turn on okay because I had the Onkyo programmed with a 1-second delay between triggers.  For example, on power-up the first trigger voltage was activated.  A second later the second trigger activated, and the third trigger finally activated two seconds after the first.  This gave each module’s signal a chance to transmit.  When the power is shut off, however, all the trigger voltages go off at once.  All three Powerflash modules send their X10 commands at the same time and they collide; consequently the appliance modules never get their signals and the amps stay on.  There’s no way to adjust for this condition in the Onkyo.  It was time for more head-scratching.  How could I delay the trigger voltages just long enough to ensure there would be no collisions?

The idea proved to be fairly simple.

X-10 Trigger Voltage Delay Circuit

Figure 1. X-10 Trigger Voltage Delay Circuit

The circuit I constructed consists of just two components, a blocking diode and an electrolytic capacitor.  The capacitor, once charged by the trigger voltage, will retain that charge for a short time after voltage is removed before dropping it.  This should be long enough for a non-delayed Powerflash to transmit its signal and avoid a data collision.  The blocking diode prevents the capacitor’s voltage from bleeding back into the Onkyo’s trigger voltage output.  Any other delayed Powerflash modules would have a different value capacitor to ensure the other  modules transmitted first.  I determined the capacitor value by trial and error.  As it turns out, a 2200 uF, 25 volt capacitor will hold its voltage in this application approximately 22 seconds.  A 220 uF cap holds its voltage for just over two seconds.  So to calculate the time delay, multiply the delay time period (in seconds) by 100, then round up or down to a standard capacitor value (in microfarads).  I built my delay circuit on a small piece of perfboard and used two mini screw terminal blocks to connect the input and output wires.  I protected the entire assembly from shorts by placing a piece of heatshrink tubing over it.

If you do any work with X10 devices, especially Powerflash modules, this little circuit could come in handy.

Leave a comment

Filed under Home automation, Projects