Metcal - GT120 Soldering System with Heater Cartridge

[gtm]

The JBC tip was a C245-407.

At what temperature?

[SteveyG]

357C, same as both Metcal units

[wickated]

The JBC tip was a C245-407.

At what temperature?

like if it mean anything in this cheapas ad. a guy that has enough money and yt chanel cant buy another tip for comparsion.
because
wow 7mm chisel solder coin very fast and 2mm chisel dont. much heat. so induction. metcal gave me 2 stations for free and some $$$ for making ad vid.

[Shock]

I mentioned this in an earlier post. According to the GT90/120 manual: "Always use cartridge removal pad and turn off system when changing cartridges."

So the part in the video at 23:30 is a bit misleading, cartridge detection does not imply hotswap capability.

[SteveyG]

I mentioned this in an earlier post. According to the GT90/120 manual: "Always use cartridge removal pad and turn off system when changing cartridges."

So the part in the video at 23:30 is a bit misleading, cartridge detection does not imply hotswap capability.

It seems to allow you to do it though, so hot-swapping works.

[SteveyG]

Power switch looks accessible but why not just put it on the front/top/side. It looks like they went with a plastic handpiece which is a bit of a backwards step in my opinion over the MX5200 with MX-H1-AV.

I forgot to mention, it's actually an aluminium handpiece that's painted black. I originally thought it was plastic too.

[thm_w]

Update video, thanks for putting all that money into this :



I wonder if they've ever considered or attempted a version with the RF heating, and having the heater slug portion removable as a swappable tip. Or if its just not possible to get good enough coupling.

[Shock]

You need to eliminate human error when testing I think. On the Metcal MX5200 your feed rate was slower so the mean external tip temp was higher. This all alters the results, it's been proven subconsciously we will skew testing by being involved especially when most people (even I) would be biased towards the MX5200 going into the test (as it's a known top performer).

As all these stations thermal profiles and tip masses are different you're comparing apples to oranges. I think you need to know external tip temp and chart the performance so you can conclude the mass of the tip. This would be done by selecting a few different sized tips in the same geometry type per station and coming up with some kind of mass per power delivery time ratio.

The Pace has a soft heating curve near set temp. JBC has probably the most aggressive one on the market. The MX5200 is likely somewhere in between but we will give it the benefit of the doubt as having most efficient power delivery, so fast and likely the curve shorter than Paces.

But I'm assuming the Metcal isn't overshooting and my point being without knowing all this ahead of time. The overshooting or worse regulating station (during recovery) with more power delivery that heats up a larger "looking" tip in a shorter time makes sense it will always be the winner.

So they all need to be tested for their recovery regulation and tip mass because with similar power deliveries and heating speeds these are the performance indicators the will be predominant.

If you don't test that then please let me send you the "worlds best soldering station" as I remove a hot iron poker from the fire. It would simply come down to the raw heating speed if you aren't considering mass and regulation near set temp.

The other thing expecting the GT120 to be fully hotswap when there is no indication of it being so. It would be nice to get a proper answer as based on other stations I would say hot plug at best (since it has cartridge detection).

Anyway nice, it's a more expensive aluminum handle than I thought and may be even be better performing than the old MX500. They just need to ditch the nut and build the supply into the station so you don't need to have a power brick on your bench. Would have better curb appeal.

[wickated]

You need to eliminate human error when testing I think. On the Metcal MX5200 your feed rate was slower so the mean external tip temp was higher. This all alters the results, it's been proven subconsciously we will skew testing by being involved especially when most people (even I) would be biased towards the MX5200 going into the test (as it's a known top performer).

at least this vid looks like real comparsion. even using jbc station cheapass clone instead of original device looks more reallife experience than ever. and perfomance for same type tips (gt120/c245) is exact same. metcal 5200 could actually perform like ~5 more watt better thermal conduction due to lack of feedback loop(tip end-heater core-thermocouple) , which is enough to grill coin better as shown in vid.
googled about 2p coin size - its 25mm 7g bronze coin, tinned in 10 secs in vid. in my country we have biggest available 22mm 6g brass coins, which are tinned in 7secs. so this vid is actually accurate.

[thm_w]

You need to eliminate human error when testing I think. On the Metcal MX5200 your feed rate was slower so the mean external tip temp was higher. This all alters the results, it's been proven subconsciously we will skew testing by being involved especially when most people (even I) would be biased towards the MX5200 going into the test (as it's a known top performer).
...

Feeding in faster could cause the tip temp to reduce more and control loop would pump more power in, giving an advantage.
I don't see him "needing" to do these tests, but they may be interesting if he really feels he has the time (dude is busy).

Anyone reasonable would measure the idle tip temperature of your hot poker and realize it was too high. All metcal/jbc/pace will have idle tip temps within ~10C of their expected setpoint.

In terms of overshoot, you can see from JBC's own 2009 measurements they can have significant ~20C overshoot. Ersa and weller are much worse (at least the models tested here). Metcal has no overshoot.
If you are doing mil-spec level soldering, sure you would care a lot.
If you are hobbyist, soldering big copper areas, most won't care about 20C of overshoot if it means the iron performs better.

[wickated]

Feeding in faster could cause the tip temp to reduce more and control loop would pump more power in, giving an advantage.
In terms of overshoot, you can see from JBC's own 2009 measurements they can have significant ~20C overshoot. Ersa and weller are much worse (at least the models tested here). Metcal has no overshoot.
If you are doing mil-spec level soldering, sure you would care a lot.
If you are hobbyist, soldering big copper areas, most won't care about 20C of overshoot if it means the iron performs better.

to eliminate human error u should joint coin pretinned with known amount of solder instead of tinning it for first time.

dropping temp would not only say to pid to pump up power but also lead to slight overshoot. for example - highend jbc stations do profiled joint solder. cpu detects temp drop speed and suggests how big area is and give as much power as needed to make good solder joint. real tip temp would even be much hugher than set temp. this works better than standard pid which actually drop power when tip temp is near set instead of giving some more.
inductive metcall actually dont need this due to lack of feedback loop and pid, but metcall instead use extremely high temp tips - 400c is like standard. this often make them look much better. and metcall tips dont burn as fast as resistive tips, so 400c inductive tips wears like 350 resistive  one.

[Shock]

My point was the importance of how mass comes into play, look at the size of the GT120 tip it's huge. The MX5200 appears to do a better job with far less mass. This test doesn't highlight the potential shortcomings (well aside from the T12 isn't so great).

It's why I think profiling the tips with temp measurement externally helps. The two most important things are temp and staying within target range. Mass (as I eluded to in my last post with the hot poker analogy) can be used as a crutch when soldering just a single target, it has a larger initial stored energy.

So recovery speed on successive joints as shown in the JBC charts will uncover shortfalls that make it hard to hide with mass. Since the GT120 seems slower than the MX5200 to heat up there evidence already it's power to mass is lower.

[wickated]

Since the GT120 seems slower than the MX5200 to heat up there evidence already it's power to mass is lower.

trick is not just power to mass, but power to temp delta. the less your temp drops, the more heat goes through. for example if i set 300c and just try to melt solid ton of copper, tip willl drain only 50 watt, but if i set 350c, power will rise to 70w. big tip internal mass means problems with heat transfer from heater to end, cuz you use copper tip material and drain surface you try to solder is copper too(so there always will be temp delta on tip end to heater core) . so fast feedback loop is better than huge mass - metcal has lower temp drop on tip end, its very little, but it makes sense.

[SteveyG]

You need to eliminate human error when testing I think. On the Metcal MX5200 your feed rate was slower so the mean external tip temp was higher.

Not sure how you've come to that conclusion or whether we're talking about different things - The feed rate was as fast as possible. The irons with the slower feed rate were not melting the solder as quickly, but the MX5200 certainly didn't have the slowest feed rate. In the earlier video the criticism was that the faster feed rates were of benefit to the heating of the coin. 

It's not a scientific test of course, and real-world soldering is rarely so demanding.

[thm_w]

New white paper on the GT90 and GT120.
Would be nice if they tested against the older stations, but, its against some "unnamed" competitors.

The test setup they used isn't terribly complicated. They use large chunks of copper, and wait until the temperature reaches a setpoint:


They say "tip or cartridge" but they obviously are using their own cartridge design to get the best performance.





The unnamed competitors, my guess in brackets:
- A) 90W tip style (Weller WTP90) Weller WT1010
- B) 75W cartridge style (Hakko FX951)
- C) 150W tip style (Ersa i-con 1) Weller WT1010H
- D) 140W cartridge style (Hakko FM-2030)


If the 140W is the JBC, then it may not be capable of 140W, in one of SDG's videos he said it limited the output to 90W. You probably have to use T470 to get the full 140W. Sadly JBC does not make this clear in their documentation, that I can find, they just say "140W peak" for the base unit.

edit: the 140W could be a Hakko too, didn't realize they had a 140W handle.

[SteveyG]

New white paper on the GT90 and GT120.
Would be nice if they tested against the older stations, but, its against some "unnamed" competitors.

I saw this too. I'm contemplating rigging up a similar setup.

[thm_w]

I saw this too. I'm contemplating rigging up a similar setup.

I can machine the copper posts. Although I'm not sure what they did to attach the thermocouple to the copper. You can see tape and some thermocouple wires, not clear where it goes. Maybe just sitting on the PCB.



For simplicity you could use a single copper block, with the thermocouple jammed in the side somewhere. Similar to a 3D printer heater block (they sell brass and copper ones).


For the tip measurement:
"Type-K thermocouples were spot welded to the soldering tip of each soldering system under test."

Kind of destroys your tip at that point. What I did was tape thermocouple to the tip, sort of half works, or just ignore tip temp and focus on the block temperature (would tell you raw performance but not overshoot).

[thm_w]

Added more details about what competitor stations were, it looks like they originally left out JBC, probably because its performance was good comparably:
https://www.metcal.com/solder-tips/top-3-factors-that-influence-hand-soldering-performance/





They show some graphs with overshoot but the GT90 is going just as much over temp, lol