Cooling Reviews Thermal grease and pads

CAYOM CA-4 vs. BOLIGO Z980 Thermal Pastes Review – More than just similar bargains?

Paste condition after the tests

The two images show the condition of the pastes after a large number of test cycles, i.e. under realistic thermomechanical loads. I can easily see how stable the inner structure of a paste remains and whether weaknesses in the dispersion or in the binder-filler ratio become apparent. The first picture shows the CA-4. The surface appears slightly roughened after the numerous load changes and shows a finely granulated texture. This indicates that the binder has retracted somewhat during the cycles and the fillers are more exposed. At the same time, however, the layer remains largely closed, with no visible cracks or open contact points. The paste shows a homogeneous displacement pattern, which indicates sufficient internal cohesion. Although I can see the first signs of incipient segregation at the edges, which may have increased due to the repeated pressure load, the CA-4 appears remarkably stable overall for an entry-level paste. Such a slightly porous surface is typical of many oxide-filled pastes and does not necessarily lead to a loss of performance as long as the contact surface remains closed.

The second image shows the Z980, which appears much more dusty and grainy on the surface. The finely distributed, almost powder-like deposits indicate that the fillers have less binding within the matrix. The surface appears matt and more homogeneously structured than with the CA-4, which indicates a stronger migration of the binder. At the same time, however, it should be positively emphasized that the layer does not show any cracks or local breakouts. Here too, the contact surface remains closed and stable. The uniform, almost powder-like texture may indicate a somewhat softer binder medium, which was more strongly displaced from the upper layers by the thermal stress. For an inexpensive paste, the Z980 nevertheless shows acceptable resistance, as neither pump-out nor visible edge delamination occur.

It can be deduced from both pictures that both pastes have a certain basic stability, even if their long-term behavior does not come close to that of high-quality industrial formulations. The CA-4 seems to remain somewhat more cohesive, while the Z980 ages more smoothly and softer. For private users, this means that both pastes will probably work for a long time in normal everyday CPU use, although differences may become apparent under extreme conditions or several cooler changes.

In future, I will also carry out standardized durability tests on an industrial scale as soon as a paste qualifies for this and provides sufficiently stable initial data. This includes defined temperature change cycles, long-term constant load tests and reproducible visual inspections under standardized conditions. Only such tests allow a real statement to be made about the ageing mechanisms and long-term reliability, which goes far beyond simple performance measurements.

Summary and evaluation

The CA-4 shows itself in the measurements as a paste with solid, but not outstanding thermal conductivity, whose strength clearly lies in its very good compressibility. The minimum achievable layer thickness of around 14 µm is extremely low and thus compensates for many of its disadvantages on the material side. Microscopy shows a fine and predominantly uniformly dispersed particle distribution, with oxidic fillers dominating, which are thermally stable but not maximally conductive. The structured surface formation after many test cycles indicates a slightly thinning binder medium without any relevant cracks or tears forming. The layer also remains homogeneous and stable in terms of behavior, which ensures sufficient internal cohesion. Overall, CA-4 is a technically unagitated, mechanically good-natured paste that scores with its low BLT and predictable behavior.

The Z980 shows a slightly higher effective thermal conductivity than the CA-4 in the measured values and confirms this with a lower Rth gradient on average. Its compressibility is slightly higher than that of CA-4, but remains within the range of similarly favorable pastes. Under the microscope, the higher carbon content and the somewhat coarser particle structure are noticeable, resulting in a duller, flatter texture. While the CA-4 tends to show a more granular surface ageing, the Z980 appears more binder displaced with a fine, almost powdery surface character. Despite these differences, the contact surface remains undamaged and completely closed even after many cycles. Mechanically, it is also stable, but somewhat less cohesive than the CA-4. Thermally, it is supported by its slightly higher bulk values, which are visible in the test stand.

A direct comparison of the two pastes shows that the CA-4 clearly demonstrates its strengths in terms of mechanical compressibility and the unusually low minimum BLT for this price range. The Z980, on the other hand, tends to have better thermal conductivity and a flatter Rth curve, which gives it a slight advantage in the temperature comparison. Both pastes remain sufficiently stable in terms of durability, but show different ageing patterns, which are characteristic of their respective filler systems.

Compared to the Arctic MX-6, both pastes are surprisingly close behind, with the MX-6 achieving a consistently better balance of compressibility, effective conductivity and Rth curve. It benefits above all from its slightly higher bulk conductivity and its well-balanced binder-filler system, which allows for thin layers and hardly loses any efficiency even with larger BLT. The CA-4 comes closest to the MX-6 in terms of low BLT, while the Z980 is thermally closer to it.

For the user, this means that both the CA-4 and Z980 offer surprisingly good technical performance for the price. Nevertheless, the MX-6 remains the most balanced choice, while the CA-4 has its own niche advantages due to its thinness and the Z980 due to its slightly better conductivity.

 

Kommentar

Lade neue Kommentare

e
eastcoast_pete

Urgestein

3,083 Kommentare 2,044 Likes

Die "kohlenstoffhaltige Oberflächenreste zwischen den mineralischen Partikeln" der Z980 könnten (Achtung: Konditional) theoretisch durch chemisch an die Oberfläche der Aluminiumoxidpartikel gebundene Kohlenwasserstoffe zustande kommen. Einige der Top Pasten wie zB die besten DowSils scheinen solche metallorganischen Bestandteile einzusetzen. Der Grund dafür ist, daß damit das Entmischen der wärmeleitenden Partikel (Hauptsächlich Aluminiumverbindungen und Salze) und der organischen Matrix (Silane und Siloxane) sehr verlangsamt und vielleicht sogar ganz verhindert werden kann*. Allerdings sind solche Aluminiumoxid Partikel mit kovalent gebundenen Kohlenwasserstoffketten an der Partikel Oberfläche eben auch nochmals teurer als "nur" einigermaßen homogen feines Aluminiumoxid.
Aber: wenn die Z980 tatsächlich solche metallorganische Bestandteile hätte, würde ich das als Hersteller lautstark verkünden.

* Aluminiumoxid ist wegen seiner ausgeprägten Polarität eher lipophob, weshalb sich diese Partikel mit den doch eher lipophilen Silanen und Siloxanen nicht so einfach beständig vermengen lassen. Wenn die Aluminiumoxid Partikel aber gebundene organische (Kohlenwasserstoff) Reste an der Oberfläche haben, ist das Problem praktisch gelöst.

Antwort Gefällt mir

Igor Wallossek

1

13,159 Kommentare 26,153 Likes

Dow macht das bisher einzig und allein über das Bindemittel. Die TC-5980 (kein Tippfehler) sollte sogar auf eine komplett Silikon-freie Matrix setzen, weil sie dann ab 80 Grad (daher der Name) wie ein Pad aushärtet.

Ich denke mal eher, der Kohlenstoff der Z980 resultiert aus einem höherverarbeiteten Siloxan. Denn man sieht es am Abrissbild auf der letzten Site, dass sich die Paste schon einigermaßen entmischt hat.

Antwort 1 Like

e
eastcoast_pete

Urgestein

3,083 Kommentare 2,044 Likes

Bei dem Z980 wär ich auch sehr überrascht (und beeindruckt) gewesen, wenn die hier doch eher "cutting edge" Materialien mit verwendet hätten. Das mit der DOWSIL TC 5980 finde ich sehr spannend.

Antwort Gefällt mir

Igor Wallossek

1

13,159 Kommentare 26,153 Likes

Die 5980 ist mehr als doppelt so lange haltbar wie das PTM7950 :D

Antwort 1 Like

e
eastcoast_pete

Urgestein

3,083 Kommentare 2,044 Likes

Darüber wird sich Honeywell aber freuen 😁!

Antwort Gefällt mir

Igor Wallossek

1

13,159 Kommentare 26,153 Likes

Die waren beim H100 schon aus dem Rennen 🙂

Antwort Gefällt mir

_
_roman_

Veteran

292 Kommentare 92 Likes

Auf Seite 5 ist zweimal diselbe Wärmeleitpaste abgebildet. Ich vermute da hätte ein anderes Bild rein sollen, oder?

Antwort Gefällt mir

Danke für die Spende



Du fandest, der Beitrag war interessant und möchtest uns unterstützen? Klasse!

Hier erfährst Du, wie: Hier spenden.

Hier kannst Du per PayPal spenden.

About the author

Igor Wallossek

Editor-in-chief and name-giver of igor'sLAB as the content successor of Tom's Hardware Germany, whose license was returned in June 2019 in order to better meet the qualitative demands of web content and challenges of new media such as YouTube with its own channel.

Computer nerd since 1983, audio freak since 1979 and pretty much open to anything with a plug or battery for over 50 years.

Follow Igor:
YouTube Facebook Instagram Twitter

Werbung

Werbung