SOLENOID RELIABILITY 2

[jamesepdavis]

BY USING ROTERS I DESIGNED A SOLENOID WITH CENTER PLUNGER AND OUTSIDE SHELL TO COMPLETE THE MAGNETIC CIRCUIT. THIS SOLENOID GETS 3 AMP PULSE AT 25 HZ. IT PRODUCES 50 POUNDS
FORCE. IT'S AN ELECTRIC HAMMER. IT'S CONSTRUCTED FROM CMI-C IRON AND HAS A COOLING FAN TO CONTROL THE TEMPERATURE.
SO FAR SO GOOD.

WE NEED THIS SOLENOID TO RUN 500 HOURS. WE HAVE A TEST STAND WHERE WE CYCLE IT AGAINST A 10 POUND SPRING. RUNS AT ABOUT 100 C AND 100 V BEAUTIFULLY FOR ABOUT 10 OR 15 HOURS. QUITS ABRUPTLY.

ELECTRICAL PULSE TRAIN IS THERE. WE THOUGHT WE HAD SHORTED TURNS IN THE COIL, BUT THE COIL IS PERFECT AFTER 15 HOURS. WE ADDED "HEAVY" INVERTER INSULATION ON THE WIRE. STILL FAILED IN LESS THAN 20 HOURS.

ONE DAY ON LOOKING AT A FAILED UNIT, I DECIDED TO CHANGE THE PLUNGER. POWER RETURNED. I'VE RETURNED TO ANNEALING THE IRON BEFORE ASSEMBLY WHICH IS DONE TO INCREASE MAGNETIC FLUX. WE HAVE PLENTY OF MAGNETIC HIT SO I HAD GIVEN UP ANNEALING THE PARTS. WILL TEST THE ANNEALED ASSEMBLIES NEXT WEEK BUT I DON'T HAVE ANY GREAT HOPE THAT THIS WILL RESOLVE THE PROBLEM.

QUESTION; IS THERE A BETTER ALLOY CHOICE TO RESIST VIBRATION AND IMPACT. THE TEMPERATURE IS ONLY 100 C, BUT THE 50 POUND IMPACT AT 25 HZ MIGHT BE CHANGING THE GRAIN STRUCTURE. THE PLUNGER DIAMETER IS ABOUT 1.0". IT LANDS ON RUBBER PADS AT EACH END, BUT SOUNDS LIKE AN AK47 WHEN IT'S RUNNING.

THANKS FOR ANY IDEAS, WE'RE OUT.

JIM (MECHANICAL)

 

[UKpete]

Jim, I've never heard of iron magnetic properties being that affected by vibration. Of course work-hardening lowers the saturation flux density and increases the iron losses (if it is carrying an ac field) but not by much, not enough to stop it working, and especially not enough to stop it suddenly which is how you describe it I think.

If the coil is ok (how do you test it? megger it to ground, check its dc resistance, how do you monitor its temperature?) and the electrical drive is ok, it sounds like a seizure. Could the plunger be seizing? It will heat up due to eddy currents at 25Hz. Can you stop it periodically to check the temperature?

If you want to rule out electrical problems, one way to do this is to fit a search coil. This is a simple coil wound with light wire and say 10% of the number of turns on the main coil. It can be placed anywhere on the magnetic circuit, even on top of the main coil - it doesn't need to be large. When you energize the solenoid, you will have a voltage induced in the search coil that you can monitor on a multimeter or scope. The voltage amplitude will be proportional to the FLUX in the solenoid i.e. proportional to the force. If the solenoid seizes you should still get a some voltage provided the coil is working ok. If you loose the search coil voltage, it is an electrical problem.

 

[UKpete]

- correction, the search coil voltage is proportional to the rate of change of flux, and the force is proportional to flux squared. If your coil is working it will still produce a changing flux which will appear as a voltage even if the plunger isn't moving.

 

[jamesepdavis]

ukpete,

YOUR EXTRA COILS IDEA MAY BE AN EASY WAY TO CONFIRM THAT THE MAIN COIL IS ACTUALLY PULSING A FIELD. WILL TRY.

JUST BECAUSE WE SEE AN AMP PULSE ON THE SCOPE, DOESN'T MEAN THE COIL IS ACTUALLY CREATING A FIELD.

TEMPERATURE IS TRACKED BY THERMOCOUPLES AND RESISTANCE RATIO.

COIL IS TESTED FOR ITS Q AND Z VALUES BEFORE AND AFTER FAILURE. LITTLE DIFFERENCE. COILS USE 26 GAGE WIRE, 1500 TURNS. AFTER "FAILURE" WE HAVE UNWOUND ALL 1500 TURN UNDER A MIRCOSCOPE. NO SIGN OF BURN OR OTHER FAILURE. SEEM LIKE NEW.

PLUNGER IS ON SHAFT WHICH HAS SUPPORT BEARINGS AT EACH END.
NO BINDING.

THANK FOR YOUR IDEAS,

JIM

 

[UKpete]

Jim, I see what you mean, there isn't much left to check. Regarding the magnetic properties of the iron, I did look up its Curie temperature (above which it is no longer any use magnetically) which is 1043K (1418°F). So I still can't see the iron properties stopping it working although there may be other opinions out there.

If you do see a voltage on the search coil, it does mean that you have real (changing) flux going through it, that's Faraday's Law. However, because the force is proportional to flux squared, a small drop in flux will give a large decrease in force. The force is also inversely proportional to airgap squared, so a small increase in airgap will also give a large decrease in force.

Perhaps you have been operating near to the edge and some thermal effect is just enough to stop it.

That is the problem with solenoid actuators, the force characteristic is very non-linear. The most effective type of actuator has permanent magnet excitation, it gives linear high-amplitude force. The force is also unipolar i.e. if you reverse the current then the force is in the opposite direction, so you don't necessarily need a spring. If you search "voice coil actuator" on yahoo you will find them.

 

[israelkk]

When you stopped do you mean the solenoid doesn't pull or the spring doesn't able ro return the plunger?

Did you check if the plunger or the housing have a large residual magnetic flux?

Did you checked the spring to see if it retained the 10lb?

How much is the gap between the plunger and the pole?

How much is the plunger travel?

Can you put a drawing or image on the net to look at?

Is the shaft magnetic or non magnetic material?

 

[dgallup]

I think you may be experiencing magnetic aging. If you re-anneal the plunger does it start working again? You could try silicon iron B-FM from Carpenter. Much more resistant to magnetic aging than electrical steels.

 

[jamesepdavis]

israelkk:

SORRY, I'M STILL TRYING TO FIGURE OUT THIS THREAD STUFF. DOESN'T SEEM TO WORK FROM MY SECOND COMPUTER.

ANSWERS TO YOUR GOOD QUESTIONS:

1. WHEN PLACED IN THE TESTSTAND IT'S PRELOADED 5 POUNDS. IT THEN ACTUATES .20" TO 10 POUNDS. SINCE WHEN "NEW" IT HAS A 55 POUND PULL AT .032", IT EASIY HANDLES THE .20" STROKE TO 10 POUNDS.

WHEN IT FAILS, IT CAN'T OVERCOME THE 5 POUND PRELOAD.

2. THE FAILED PLUNGER HAS NO MAGNETIC QUALITY. IT WON'T PICK UP A PIN.

3. ACTUALLY THE "SPRING" IS A TENIS BALL FOR DAMPING. IT'S STILL GOT ITS F/X.

4. GAP IS 062" SILICONE PAD. SO THE REAL STROKE IS .062 + .2".

5. SHAFT IS NON-MAGNET STAINLESS.

6. I DO HAVE A SECTION OF THE ASSEMBLY ON A .DOC PAGE.
I WOULD NEED YOUR COACHING ON HOW I MIGHT PUT IT ON THE WEB FOR YOU TO SEE. I COULD E-MAIL IT TO YOU. PERHAPS YOUR E-MAIL ADDRESS IS ON THIS TIPS SITE. I'LL LOOK AROUND.

dgallup and al hands....

I'M GOING TO TALK THE PEOPLE AT CARPENTER AS SOON AS I GET SMARTER. YOUR INPUTS ARE HELPFUL IN OPENING NEW IDEAS.

CURRENTLY ALL THE SOLENOID PARTS ARE AT CMI AND THEY ARE RUNNING MAGNETIC PROPERTY TESTS. WHEN I GET THEIR CONCLUSIONS I'LL GET THE ANNEALED PARTS REASSEMBLED AND RESTART TESTS. IT'S QUICK TO GET THRU 20 HOURS. IF WE'RE STILL FAILING I'LL MOVE ON TO CARPENTER AND DISCUSSION OF A BETTER ALLOY.

I REALLY APPRECIATE EVERYONES INPUT. STAY TUNED. I'LL REPORT CMI'S CONCLUSION. WE'LL GO FROM THERE.

THE WEATHER IN CALIFORNIA IS INDUCING THUNDERSTORMS FROM THE SOUTH AND THAT MEANS GOOD SOARING CONDITIONS. IN REAL LIFE I'M A CROSS COUNTRY GLIDER PILOT. SO IF I GO SILENT FOR A FEW DAYS I'M JUST ON RETRIEVE.

JIM

 

[jamesepdavis]

ALL HANDS,

AFTER TWO ANNEALING PROCESSES, FINALLY GOT SOFT PARTS.
RE-ASSEMBLED AND TESTED. FAILED IN 10 HOURS, SO WAS A WASTED MONTH.

TODAY, FINALY GOT TO TALK TO METALURGIST AT CARPENTER.
HE SAYS THAT IRON CORE PRODUCTS LOSE THEIR PROPERTIES WITH AGE. IF YOU HAVE A SOLENOID VALVE THAT'S ONLY OPERATING TWICE A DAY YOU THINK THESE MATERIALS LAST FOREVER. IF YOU HAVE A SOLENOID THAT OPERATING AT 25 HZ FOR SEVERAL HOURS EVERY DAY, YOU'LL FIND THAT PURE IRON AGES AND FAILES THE MAGNATIC CIRCUIT. HE AGREES WE SHOULD USE CARPENTER CORE IRON B-FM AND IS ARRANGING A SAMPLE SHIPMENT.

MAY NOT LAST FOREVER, BUT MAY GIVE US ENOUGH PRODUCT LIFE.
WILL TEST. ONE TEST IS WORTH A THOUSAND EXPERT OPINIONS.

ALL THE BEST,

JIM

 

[dgallup]

We use silicon core iron B-FM & various magnetic stainless steels in fuel injectors. Design life is 600,000,000 cycles, we have tested to over 2 billion cycles, no aging problems. You could also look into 430F(R) or Chrome Core 13-FM but you will loose some magnetic force if you are running up near saturation.

Before the invention of silicon iron, transformers were made with electrical steel laminations. It was common practice to disassemble the transformers periodically to re-anneal the core laminations.

 

[jamesepdavis]

dgallup,

THANKS FOR YOUR INSIGHTS ABOUT B-FM. CARPENTER IS SENDING A SAMPLE OF OUR SMALL DIAMETER NEXT WEEK. A LARGER DIAMETER
SAMPLE WON'T BE AVAILABLE UNTIL END OF DECEMBER.

AS YOU KNOW CARPENTER ONLY BUILDS TO ORDER. IS THERE ANY OTHER SUPPLIER OF SILICON IRON ON THE PLANET? I'VE BEEN DOING GOOGLE SEARCH BUT FOUND NOTHING SO FAR. HAVEN'T GIVEN UP.

THANKS AGAIN FOR YOUR GOOD HELP,

JIM

 

[UKpete]

Jim

I have been following developments with your problem but not understanding what is going wrong. I still find it difficult to believe that any aging effect will be causing it, because unless someone can convince me otherwise (e.g. with published data) I don't think the magnetic properties can change that much without some rather extreme heat treatment that may alter the crystalline structure.

Work hardening (treated by re-annealing) does affect magnetic properties of materials but mostly in terms of increasing the hysteresis loss, not by decreasing the force capability, I think.

If I may correct you on a couple of your comments:
1. In one of your earlier posts you say that "the failed plunger has no magnetic properties. It won't pick up a pin". Because it won't pick up a pin it doesn't necessarily follow that it has lost its ferromagnetic properties, it could be waiting to be magnetized.
2. There are billions of relays and actuators out there that work continuously, not just intermittantly, without any degradation over years of operation. I've never come across one that did.

Have you tried using a plain mild-steel plunger just to see what happens? You may need to increase the current to get the same force. I challenge anyone to take a heavy hammer to a bar of mild steel for as long as you like and make any significant impression on its bulk magnetic properties!

 

[jamesepdavis]

dgallup,

I THOUGHT YOU WERE SUPPORTING THE CARPENTER METALURGIST THEORY ABOUT "MAGNETIC AGING". I THOUGHT YOUR COMMENT ABOUT 600,000,000 CYCLES SUPPORTED B-FM RESISTANCE TO SUCH.
I WAS GOING TO ASK HOW THAT DATA CORRELATED TO OUR EXPERIENCE. I AGREE THAT I HAVEN'T SEEN ANYTHING TO EXPLAIN WHAT'S GOING ON.

WE'LL HAVE A PIECE OF B-FM A WEEK FROM TOMORROW. WE'LL MAKE A PLUNGER.

WHEN I "PIN" TESTED A FAILED PLUNGER I WAS LOOKING TO SEE IF IT HAD DEVELOPED A STRICK POLARITY. NONE.

I HAVE A FAILED UNIT SITTING ON MY BENCH. I HAVE NEWLY ANNEALED PLUNGER TO DROP INTO IT.
WILL DO TOMORROW.

I'VE BEEN CALLING TO FIND ALTERNATE SOURCE OF SILICON CORE IRON. IS CARPENTER THE ONLY SOURCE? THEY REQUIRE A MILL RUN WITH SIX MONTHS DELIVERY.


JIM

 

[dgallup]

Carpenter is the only source for Silicon Core Iron B-FM. Several other places make non-free machining grades of silicon core iron but only in thin sheets for transformer laminations. You will have to go to 430FR or similar if you want other suppliers. Ugine is an alternate source for 430FR.

 

[jamesepdavis]

ALL HANDS,

YOU'LL RECALL THAT NO ONE WAS CONVINCED ABOUT "MAGNETIC AGING", BUT I AGREED TO PURSUE CARPENTER BFM ANYWAY.

I FOUND THE ONLY 2.5" DIAMETER, 6" LONG PIECE OF BFM AVAILABLE ON THE PLANET AT SCIENTIC ALLOYS. CARPENTER WON'T HAVE ANY UNTIL AFTER THE FIRST OF THE YEAR.

WE MACHINED ALL NEW SOLENOID PARTS FROM BFM. ANNEALED AT NO SMALL COST. RE-ASSEMBLED. RAN LIFE TEST AT HALF POWER INSTEAD OF FULL POWER. THE UNIT FAILED IN 45 HOURS.

AT THE OUTSET THE BFM MATERIAL GAVE A NOTICEABLY SHARPER HIT. IT ALSO RAN 10 % COOLER. AFTER 45 HOURS, THE SOLENOID HIT DROPPED FROM 50 POUNDS TO LESS THAN 10 POUNDS.

I THINK THERE'S MORE THAN ONE THING WRONG SO AS WE APPLY OUR LOGIC AS THOUGH THERE IS ONLY ONE PROBLEM, NOTHING MAKES ANY SENSE. I'M GOING TO SPEND SOME TIME PROBING CIRCUITS AND LOOKING AT SCOPE PICTURES TO SEE IF I CAN GET A CLUE. I HAVE OTHER COILS AND PLUNGERS THAT CAN BE INSTALLED. I'M SURE IF I DROP IN A NEW PLUNGER, POWER WILL BE RESTORED. I'LL THROW IN THE EXTRA COIL THAT WAS SUGGESTED EARILER.

ANY IDEAS WILL BE APPRECIATED. I'M OUT.

JIM

 

[jamesepdavis]

TO ALL HANDS WHO KINDLY RESPONDED TO MY QUERY.

YOU WERE RIGHT THE FIRST DAY. IT WAS THE BEARINGS.

I WAS RUNNING THE UNIT ON ITS SIDE IN THE TESTSTAND SO ALL THE BEARING LOAD REMAINED IN ONE SPOT. IF YOU PICKED UP THE COLD SOLENOID IN YOUR HAND YOU'D SAY "WHAT BEARING WEAR?" BUT RUNNING CAUSE OVAL BEARING WEAR AND BIG DRAG ON ONE SIDE. HEAVY STICKION WAS LOCKING THE UNIT UP.

REPACED THE TEFLON BEARING WITH SINTERED BRONZ OIL IMPREGNATED. PUT THE TESTSTAND IN THE VERTICAL POSITION, WHICH IS TYPICAL OF ITS USE, NOW HAVE RUN BEYOND 200 HOURS ON TWO UNITS.

SINTERED BRONZ BEARING ARE TOO SOFT. WILL REPLACE, BUT IT WASN'T SHORTED TURNS OR CHANGED IRON GRAIN STRUCTURE. I BUILT A COMPLETE SILICON IRON SOLENOID. IT ONLY RAN 10 HOURS UNTIL THE ABOVE FIXES. ONCE WE STARTED RUNNING IT SIDE BY SIDE WITH THE PURE IRON IT RAN 15 DEGREES C COOLER THAN THE PURE IRON UNIT. BOTH HAVE RUN OVER 200 HOURS.

PROBABLY GOING TO USE PURE IRON IN PRODUCTION BECAUSE IT'S IMPOSSIBLE TO GET SILICON IRON FROM CARPENTER STEEL WITHOUT BREAKING THE BANK AND THE SCHEDULE.

THANKS AGAIN FOR YOUR HELP,

JEPD

 

[UKpete]

Jim, that's good news. Can you get away with using mild steel? The only advantage of silicon steel is that it's got higher resistance so the eddy current loss is reduced. If eddy currents are a problem you should be using laminated steel anyway, not solid.

 

[jamesepdavis]

I NOW THINK YOU COULD USE ANY STEEL LIKE L1214. THE BIG ATTRACTION OF SI-I IS THAT THE WHOLE SOLENOID RUNS COOLER
BIG TIME. THE COIL DOESN'T GET AS HOT, THE COIL RESISTANCE STAYS LOW AND THERE IS NO DEGRADATION IN HIT POWER.

I'M GOING TO HAVE A RUN AT CARPENTER TOMORROW. I TALKED TO THEM AT A DESIGN SHOW AND THEY WERE NOT ENCOURAGING.
BEST ALTERNATE IS CMI-C PURE IRON READILY AVAIABLE AT $2
A POUND.
THANKS AGAIN FOR YOUR HELP.

JIM

 

[dgallup]

430FR has even higher resistivity than silicon iron and is frequently used in solemoid applications. Saturation flux density is reduced so you will get less force from the same geometry. Availability is not too bad but cost is higher.

 

[jamesepdavis]

dgallup,
per your earlier suggestion i pursued 430fr. it was as hard to get as bfm and had the flux disadvantage. the cmi-c is cheap and has quick delivery. my testing paid off. we can use cmi-c and go into the sunset.

jim