Zoals jullie weten “vecht” ik al jaren tegen candy stuwstof met lage dichtheden welke veroorzaakt worden door minieme gas insluitingen, de zogenaamde microbubbles. Om dat tegen te gaan had ik een compressie mal ontworpen waar ik dichtheden van 99% haal in mn 54mm motoren. Met de Thrust motoren werkt het ook aardig maar is de mal niet geoptimaliseerd waardoor er 92% dichtheden gehaald worden. Echter, inmiddels weet ik dat de oorzaak ligt in de (lage) kwaliteit van het kaliumnitraat.
In het kaliumnitraat zitten onzuiverheden in. Te weten: kaliumwaterstofcarbonaat (kaliumbicarbonaat). Dezelfde stof zit o.a. in zelfrijzend bakmeel en na verhitting komt er CO2 vrij wat voor een luchtige cake zorgt. Candy stuwstof verhitten we ook en vanaf 100graden komt er dus CO2 vrij wat voor luchtige stuwstof zorgt – een ongewenst effect. Dit heb ik getest door aan ons kaliumnitraat een beetje natuurazijn toe te voegen waarna er bubbles vrijkwamen.
Zaak is nu om nieuwe KN te bestellen (met een lage pH waarde) maar gezien de laatste ontwikkelingen in Noorwegen is het wellicht verstandig om dit nog even uit te stellen. Dit is voorbeeld: http://www.haifa-group.com/
Verder nog wat experimenten gedaan om te kijken of een silicone olie met lage viscositeit (DOT 5 remvloeistof) de oppervlaktespanning van de luchtbubbles kon verkleinen / verbreken. Behaalde resultaat was anders dan verwacht. Geen hogere dichtheid dan normaal via smelten onder vaccum maar het de gesmolten slurry goot veel beter. Test moet uitwijzen of het geen invloed heeft of de verbranding.
Dit was het weer even.
Groet,
Jurriaan
All batches were 200gr KNDX. Mass of inhibiter was used to determine actual density.
Test I – Normal.
Heated untill all KNDX turned into a nice slurry @ 165degr setting. Continued heating the slurry for 5min. Stirred at 3min. Final heating and stirring close to caramelisation and casted. Slurry casted normal.
Grain after being squared on the lathe. Length = 70,01mm. Mass = 180,3gr. = 85% density
Test II – Vacuum.
Heated untill all KNDX turned a nice slurry @ 165degr setting. Applied vacuum for 3min @ -0,93bar. Released vacuum, gave it a stir all took in total 1 min and applied vacuum again for 1 min. Total time heated similar to test I, also 5min. Final heating and stirring close to caramelisation and casted. Slurry casted slightly less easy then test I
Grain after being squared on the lathe. Length = 68,86mm. Mass = 191,0gr.= 92% density
Test III – Vacuum + DOT 5 silicon brake fluid (similar to Polydimethylsiloxane anti-
Heating untill all KNDX turned into a nice slurry @ 165degr setting. Inserted 0,4gr (0,2%) brake fluid and stirred abit. Applied vacuum for 3min @ -0,93bar. Released vacuum, gave it a stir all took in total 1 min and applied vacuum again for 1 min. Total time heated similar to test I, also 5min. Final heating and stirring close to caramelisation and casted. Slurry casted much better then test I or II. More like thin yoghurt. Also residue KNDX in heating vessel slid from bottom to lower half of the pot like rain drops (like heating vessel was teflon coated).
Grain after being squared on the lathe. Length = 75,1mm. Mass = 206,3gr. = 91% density.
Summery:
- This KNDX mixture produces lots of microbubbles. The KN was tested with the vinegar test and showed foaming thus potassium bicarbonate (KHCO3) is present (decomposes at ~ 100C to liberate CO2).
- Vacuum gives a significant reduction of air in the molten slurry but the efficiency is heat sensitive and depending on layer thickness.
- The discrepancy between test II & III could be addressed to:
- The heating thermostat which comes on an off within the time limit of 5min of heating. It was clearly observed that when the heating vessel was on, air started vividly bubbling out of the molten slurry and reduced when the heating vessel switched off. It could be possible that test II had more energy / heat put into the molten slurry than test III thus releasing more air.
- More of the top of the grain was machined off during squaring which resulted in a shorter grain. This removed top layer is known for inclusion of small bubbles which are formed as the remainder of the KNDX from the heating vessel is scraped together and having a lighter density then the rest of the grain.
- Adding DOT 5 silicone brake fluid to the mixture of molten KNDX did gave a significant viscosity reduction and eased the pouring of molten KNDX.
Dot 5 mostly consists of Polydimethylsiloxane. Minimum of 70% but usually 95%
(C2H6OSi)n
http://en.wikipedia.org/wiki/
centistokes: 1 cSt = 0.01 St = 1 mm2/s
1cSt ➝ 1mm2/s ➝ 1 mPa.s
42.7cSt |
Viscosity at 100°C 6 mm2/s =