T12 clone vs Metcal SP200

[mzzj]

Did a quick comparison between two of them and I have to say that I was expecting T12 clone (stm32) to be better.
Seems like the thermal coupling between tip and the thermocouple is the weakness at least with these T12 clone tips.

T12-BL on clone and almost identical sized tip SSC-722A on metcal. Both iron tips verified to be at 360Cel +-5cel , Metcal does about twice the amount of solder joints in same time.
T12-K tip on clone is about par with the metcal but the tip is huge chisel tip.

Looks like there is not much other benefit than faster heat-up at higher voltages than 20 volts as even then large T12-K chisel tip runs max 60% duty cycle when attempting to solder large copper plane at 370c setting.

Are the genuine T12 tips better or is the sensing element as far from the tip as in the clones?

[GreyWoolfe]

I have both a Hakko FX-951 with T12 tips and a Metcal MX-500.  I prefer the Metcal.  The wand feels better for starters, however both silicone cords seem equally flexible.  I have not done any quantitative testing, but the Metcal seems to solder faster and better, especially large solder joints.  I still use and enjoy the Hakko but Metcal rocks.  The MX series is the 13.56 MHz model and the SP200 is the 470 KHz model.  I have not tried any of the lower frequency models so I can't offer anything there.  The smartheat system is also excellent.  No knobs to fiddle with.  Buy the tips in the temperature range you want and you are good to go.  I use the 600 series tips which is 650 F.  I have 1 800 series tip for lead free solder.  Hope this helps.

[thm_w]

Did a quick comparison between two of them and I have to say that I was expecting T12 clone (stm32) to be better.
Seems like the thermal coupling between tip and the thermocouple is the weakness at least with these T12 clone tips.

Yeah, theoretically thats what the 'cal' tip menu in some of the T12 clone firmware should be able to fix. But its still not going to be as good as a sensor right at the tip.
I'm not sure if Hakko does anything to offset this effect in the construction of the tip, or if they just expect you to bump up the temperature.
Thats kind of the advantage of the metcal, in that the calibration is built into the tip.

Are the genuine T12 tips better or is the sensing element as far from the tip as in the clones?

I don't have two identical styles to compare, but I have some small size types and the performance is similar enough.
However, mechanically the T12-JL02 clone tip I have is not very strong (one of the smallest available tips). I suspect this is due to cheaping out on the materials being used.

[mzzj]

Yeah, theoretically thats what the 'cal' tip menu in some of the T12 clone firmware should be able to fix. But its still not going to be as good as a sensor right at the tip.
I'm not sure if Hakko does anything to offset this effect in the construction of the tip, or if they just expect you to bump up the temperature.
Thats kind of the advantage of the metcal, in that the calibration is built into the tip.

Measured "unloaded" tip temperature(with k-type thermocouple) and T12 display were close enough in this case.
AFAIK the "cal tip" just applies correction for the temperature "seen" by the sensor but it's able  to compensate only "unloaded" tip temp.
For real tip temperature under use it won't help much as the sensor is not at the tip.

[Shock]

The T12 knife is a 70W tip, the Metcal is 35W. You were probably seeing the power limitation of that particular Metcal station. The Metcal MX5200 has 80W into a single channel.

The T12 isn't the highest performing cart out there by a long shot, it's slower and even 70W is relatively low power given alternatives.

[thm_w]

Measured "unloaded" tip temperature(with k-type thermocouple) and T12 display were close enough in this case.
AFAIK the "cal tip" just applies correction for the temperature "seen" by the sensor but it's able  to compensate only "unloaded" tip temp.
For real tip temperature under use it won't help much as the sensor is not at the tip.

Yeah its not going to be ideal, but the idea is under load you have the correction factor applied, so the end of the tip temperature is closer to what it should be.
As you've measured the tip unloaded is at set temp, so it will be too hot once the correction factor is applied (reduced tip life, more burning of flux, etc.).

I think the geometries could be improved slightly, they seem to taper the metal section aggressively. Shorter and stubbier the better I think.

[mzzj]

The T12 isn't the highest performing cart out there by a long shot, it's slower and even 70W is relatively low power given alternatives.

70W is plenty of power for the T12 design. T12-BL tip set to 360 Cel never goes above 30% duty cycle during use so 25 watts heater would work as good for that.
I have to use the  T12-K large chisel set to 440 Cel! to see anywhere near full power during use.

[KL27x]

Yeah its not going to be ideal, but the idea is under load you have the correction factor applied, so the end of the tip temperature is closer to what it should be.
As you've measured the tip unloaded is at set temp, so it will be too hot once the correction factor is applied (reduced tip life, more burning of flux, etc.).

By this description, using a "correction factor" should be no different than turning up the set temp. It just means you don't have to change set temp when changing tips. (Instead, you would have to change the tip setting, unless the station can automatically figure out which tip is installed.)

I have to use the  T12-K large chisel set to 440 Cel! to see anywhere near full power during use.

i think you could potentially reach higher duty cycles at lower set temp while actively soldering to larger joints/planes?

IMO, the thing at which cartridge style irons really shine (whether metcal or thermocouple doesn't really matter), is in thermal isolation between the tip and the handle of the iron. When you are making lots of joints that have huge heatsinking, the handle won't warm up so much. You can also use different rating heaters for different tips. The T12 spatula tip comes to mind. I doubt this thing uses the same power heater as the other T12 tips.

The "sensor closer to the tip advantage" is basically nonsense. The thickness of a soldering iron tip at the base is large enough that this distance is nowhere near to being the weak link. That amount of copper is a good enough conductor of heat that it doesn't matter where the sensor is. It's the part of the tip that is thinner, nearer the point, where the thermal gradient is occurring when the tip is applied to a relatively colder thermal mass and/or a heatsink. If you could get the sensor in the actual pointed part of the tip, that would make a difference. But the sensor would have to be micro-small else you're gonna lose performance by reducing the amount of copper there. There is no iron where the sensor is anywhere near the point of the tip.

Even metcal tips, this is not the case. And this is why there's a rather large range of temp in a given metcal range of tips. The skinny/pointy tips are tuned to a higher setpoint than the thicker/bluinter tips to compensate.

 

[thm_w]

I can't recall if someone has cut open a JBC tip to see where the sensor is positioned.

Metcal is a bit different as the "sensor" is the entire curie slug. I'm not sure if that slug extends into the copper tip at all, or it is just back from the edge before the tip begins to taper.

https://www.eevblog.com/forum/reviews/metcal-sttc-soldering-cartridgetip-teardown/

[KL27x]

^AFAIC, the pic #2 shows the Curie alloy slug doesn't start until the tip reaches the full thickness of the base. If you made the fine point out of Curie metal, it better have super high thermal conductivity to work worth a damn. Reading temp at the very tip does you no good if you can't transfer heat there.

If this was as the marketing department allows the average consumer to believe, then they would not have to make the setpoint of all of their "700F" tips as well higher than 700F, with the finer ones 40+ degrees hotter than the fatter ones.

[mzzj]

T12-BL tip set to 360 Cel never goes above 30% duty cycle during use so 25 watts heater would work as good for that.  I have to use the  T12-K large chisel set to 440 Cel! to see anywhere near full power during use.

i think you could potentially reach higher duty cycles at lower set temp while actively soldering to larger joints/planes?
 

Tested it already on a pieces of unetched FR6 with full copper coverage and on a large 6mm thick copper plate that barely gets above room temperature.
T12-BL wont get past 30% duty cycle if set to 360cel. (other than during start)

[Shock]

Tested it already on a pieces of unetched FR6 with full copper coverage and on a large 6mm thick copper plate that barely gets above room temperature.
T12-BL wont get past 30% duty cycle if set to 360cel. (other than during start)

Did you measure the contact area and mass of the thermal bridge you made with solder? Just kidding don't do that.

Seriously as I mentioned the T12 tip comparatively is not the best cart out there, this is across the range not just the 0.4mm conical, you already proved it was inferior to the Metcal anyway so there is no reason to go too deep into it.

The only thing you are getting with a T12 clone is a lower price and better UI (with source perhaps?) and hopefully similar performance to the FX-951. Everything else is some degree of worse than the Hakko FX-951. Not that I'm recommending going and getting a FX-951, Pace ADS200 is way better, but the T12 clones are cheap for a reason.

[bd139]

Metcal PS900 here. The quicko T12 I had was vastly inferior to even this bottom end Metcal even after I boosted the 12V source up to 24V. Plus the T12 blew up. I needed a portable DC iron for antenna work so ended up buying a 12V Weller TCP which is superior to the T12 as well and cost me about the same new in box off eBay.

[GreyWoolfe]

Metcal PS900 here. The quicko T12 I had was vastly inferior to even this bottom end Metcal even after I boosted the 12V source up to 24V.

Asking EEVBlog members which soldering equipment to buy is like asking ham radio operators which radio to buy.  Ask 10 hams and you will get 15 different opinions.  Ask 10 EEVBlog members and you will get 30 different opinions because we own multiples of everything!   What I didn't mention in my earlier post, if the deal on the Metcal MX-500 would have come along before buying the FX-951 and I had used it for awhile, I would not have purchased the Hakko.  I would have kept my 936 as a second station and maybe picked up a couple of more tips for the Metcal.

Even metcal tips, this is not the case. And this is why there's a rather large range of temp in a given metcal range of tips. The skinny/pointy tips are tuned to a higher setpoint than the thicker/bluinter tips to compensate.

Just for curiosity sake, I took out my Fakko FG-100 and measured a .8 mm chisel tip and a 3.2 mm chisel tip.  The .8 measured 334C\633F and the 3.2 measured 332C\630F.  Not THAT big a difference.  These are 600 series tips.

[Shock]

Just for curiosity sake, I took out my Fakko FG-100 and measured a .8 mm chisel tip and a 3.2 mm chisel tip.  The .8 measured 334C\633F and the 3.2 measured 332C\630F.  Not THAT big a difference.  These are 600 series tips.

The 600 series is based somewhere around 357C/675F. But temps are meant to vary across the series anyway so your results are normal. The repeatable temp alone is +/- 5C and 41F so they aren't specifically designed to be super accurate, more about recovery speed.

[Tarloth]

... so they aren't specifically designed to be super accurate, more about recovery speed.

But it's really important the "accurate" in a solder iron? If I have 310 or 330 centigrade degrees it's important? I really doubt that, I accept that its not ok a difference of 100 degrees, but 10% it's ok. The temperature drops a lot when tip contact the pad, more  than 10%. I think that insane accuracy it's a selling argument. I think that it's more important the termal mass than temperature accuracy. I hardly ever change the temperatute in my stations, recommended temperature work's perfectly and only change it when need accumulate more heat for bigger components than usual. Then the temperature isn't important at all, only matters that melts solder and not pass an excesive heat to the component.

[Shock]

To melt solder no, but if you're soldering to IPC standards I'm fairly sure +/-10% is a fail.

One of the advantages of cartridges is improved heating accuracy. If you just put minimal plating on a low mass tip and crank the heating cycles up, everyone will think it's great station until they discover the dips and overshooting. A different story on a low end station but if I was paying $20-30 a cart and hundreds of dollars for the station, I'd expect quality, accuracy and performance.

[Tarloth]

I not agree.

IPC standards refers to Peak temperature or limits. For eutectic SnPb (to use one example of low temperature) IPC recommends for most of components 220 degrees (235 for some very small components), but the eutectic point it's 188 degrees. For wetability, intermetal formation and tensile strength its better to use almost ten or 15 degrees over this temperature as to solder with a "perfect iron" with perfect heat transfer to the PCB and component.

Actually the IPC recomendations are a pair of time-temperature parameters and at temp limit it's more important the time than the accuracy of temperature, but I focus only in temperature.

If you have a infinite thermal mass iron you can solder perfectly at 200 degrees and you still are 10% away of peak recommendations of IPC. The problem it's that the solder iron hasn't a infinite thermal mass then the temperature fall from a practical point set in the iron to a medium temperature in the component meanwhile the solder melts and solidify.

The IPC specify the max temperature of the component and not of the iron. In a reflow oven the temperature it's uniform, not in an iron tip.

If you have a solder station of insane accuracy capable of stay the tip temperature between 0.001 degree but when touch a component the temperature decay 25 degrees it's really an accurate station? Was in any moment important the accuracy? iron tip without the PCB contact can mantain a very precise temperature, but the important it's that the final temperature when solder stay in margins and not overshoot in any case. Most of professional solder station low the temperature of the tip more than 15 degrees when touch a medium massive component for sure, and if not believe me, put a TC and meassure it.

If your tip has a very very good thermal mass (i.e. the temperature stay stable in any case), you can fix your sation to any temp between 200 and 230 degrees and the IPC limit it's not overshooted in any form and any time. You have a 10% range of temp for sure.

If you have a more real thermal mass, you need to set the station to any between 230-300 degrees and solder relatively fast to protect the core of component of overshoot 220 degrees. The percentage here it's near 25%!

Most of the people that solder PB free fix therir stations between 260 and 330 degrees when the IPC limit it's 260 degrees! It's really important if the Metcal, Hakko or Pace tip stay to 310 or 330 degrees in this case? At which temperature set your station?   

At the end of the day, was important if your solder station was less of +-10 degrees apart of the theoretical temp?

[thm_w]

Well the 600 series is stated as 675F (357 C) max temperature. The 7 series is 775F (412 C). So that is way above any of numbers you've given.

Metcal says "IPC J-STD-001 compliant" whatever that means.

Might have to get one of those tip testers to try it out.

[KL27x]

From Thermaltronics website (since I can't find this info on metcal/Oki website):

700 Type (350°C - 398°C)

48C/86F  spread

Yeah, I dunno how much of this is variation/tolerance vs intentional range/spread. I would assume most of this is intentional across difference tip shapes. I routinely adjust the temp on my stations based on the tip type.

As Tarloth says, there's always temperature gradient created when you are actually soldering. And the shape of the tip will have a lot to do with how big this gradient is (since the sensor simply isn't anywhere near the work contact surface of the iron, no matter how you want to interpret the brochure). A long skinny tip is gonna sag. That small cross section of copper is gonna have "thermal resistance" and you will get temp drop across is, same as voltage drop across a resistor.

[Shock]

I accept that its not ok a difference of 100 degrees, but 10% it's ok. The temperature drops a lot when tip contact the pad, more  than 10%.

At the end of the day, was important if your solder station was less of +-10 degrees apart of the theoretical temp?

I noticed your percentage turned into degrees somewhere over several paragraphs, next I expect you to turn Celsius into Fahrenheit to argue your point.

Last time I checked many standards were around +/-6C or lower. If you think your 10% accuracy story is acceptable go argue with the whole industry, quote IPC scripture mixed in with your opinion for heightened effect!

[Tarloth]

I never use Fahrenheit and I did could use +-10 degrees because in theoretical 200 degrees that represent a 10% of variation. There are no inconsistencies in the argument, please post exactly where you found any, maybe I'm expressing badly in english, sorry.

I repeat, IPC standards define a MAXIMUM temperature with a thin margin, not the solder exact temperature at any time. Please, post the IPC text where they "dictates" that the component would be solder at this exact temperature  and not less or more.

And a common sense question.... Why Metcal sell three series of tips at different temperatures that all guarantees IPC standard? Why All the solder stations that comply with IPC standards allow freely change the temperature?

I answer it for you. Please read this IPC standard (I took the first that google sugest to me)

http://www.ipc.org/4.0_Knowledge/4.1_Standards/test/2.4.36b.pdf

If you read with attention the condition of +-6 degrees (I use always centigrade!) in 4.1 it's to stablish the accurate of the solder station that you use in the test. BUT IPC say's that the experiment can be done using three different temperatures 260,315 and 371 centigrade degrees...WHAT?

In point 5.4 IPC affirm "If the solder connection cannot be formed within five seconds using the specified method the test shall be repeated using the subsequent method. This deviation shall be noted in the test report". That is that IPC asumes that if the thermal transfer it's not sufficient (by the pad mass or tip) you need to use a temperature 50 degrees high!!!!

But if this text it's not enough in point 6.2 affirms that if you use a solder station with fix temperature tips like Metcal you can use for the 260 c degrees option the 500 series that have a max temperature of 302 degrees!!!! IPC asumes that you CAN solder at 260 degrees or 300 and it's the same experiment and both are equivalent!!!!!!!

IPC SAYS (not me!) that it's EQUIVALENT use 260 degrees or 300 degrees, 40 degrees of difference, a rough 15% upper of the "recommended" temperature

By the way, in the same 6.2 point 315 c option recommends 600 series with max temp of 357 c and for 371 option you can use the 700 series with 412 degrees!!! (https://www.wsbenelux.nl/wp-content/uploads/2014/11/metcal-tip-temperatures.pdf). In all cases IPC assumes that a difference of 40 degrees ITS EQUIVALENT AND ADMISIBLE

The specification of +-6 degrees it's only for repetibility of test between stations, to compare results, but in the same point (4.1) affirms without  doubt that same solder station can be replaced by a metcal with 40 degrees more in each option!!!

But let's me read that some other big companies talk about this point. JBC recommends 300 to 350 c degrees (depends on component size)  for a lead free solder that IPC marks a max of 260 C degrees  and melt at 220 c degrees. ( https://www.jbctools.com/blog/correct-temperature-soldering/)

AGAIN, a 40 c degrees difference seems to be ok!!!!

When you solder EVERY day, which temperature selects in your station?

[KL27x]

When you solder EVERY day, which temperature selects in your station?

Mine Hakko 888 lives at around 300-305C, 95% of the time, for leaded solder.

Even though the solder melts at under 200C, the temp sensor of the station is regulating the back of the tip, not the joint.

If you could make an iron with infinite power, sensor in the very point of the tip, and ridiculous accuracy using ESP, you would still need to set the temp higher than the melting temp of the solder. Else you would have an iron that could maybe melt its way through a blob of solder without reflowing the entire blob.

[Shock]

I never use Fahrenheit and I did could use +-10 degrees because in theoretical 200 degrees that represent a 10% of variation

I'm not great with maths but isn't 190C and 210C a 5% variation of 200C? Anyway, despite me wanting to ridicule you over the using 200C as an example, please lets just stop.

The whole idea of accuracy is conforming to an easy to interpret standard, you're not helping anyone nor will get anywhere by arguing against it. Pace has had +/-1.1C tip accuracy for many years. The tips are $11 the station is $200 so its not like you are paying anything extra for increased accuracy so why argue against it?

[Tarloth]

Yo may say what you want. The information is out there in the documents, the temperature used in your iron it's OK if the core temperature of the component nor exceed the stipulated by the maker of the component or the IPC norm. The norm it's a limit, and applies to refusion oven with an specific curve and well characterized.

If you try to solder some component with your iron and the solder not melt, you must increase the temperature to improve the heat transference. This procedure it's universal and the IPC norm specify it too.

The accuracy of the station can be fixed with a simple external thermocouple well coupled to the tip that measure the temperature exactly at the extreme of the tip. At the end of the day you need an external device that confirm what your station it's reading, but a difference of ten centigrade degrees it's practical negligible when the IPC itself recommends modify the temperature in 40 centigrade degrees when the component can't be correctly solded.