iTECH RF-Axxx reflow oven algorithm is outdated

[F1GWR]

Beware! These reflow ovens only incorporates an old-style RSS (Ramp-Soak-Spike) algorithm. More efficient and safer RTS (Ramp-To-Spike, google this term), although easier to program in firmware, is not implemented. When asking to the company, they simply ignore the request and stick to RSS.

Did anyone develop a RTS firmware for these RF-A series?

[SMTech]

Beware! These reflow ovens only incorporates an old-style RSS (Ramp-Soak-Spike) algorithm. More efficient and safer RTS (Ramp-To-Spike, google this term), although easier to program in firmware, is not implemented. When asking to the company, they simply ignore the request and stick to RSS.

Did anyone develop a RTS firmware for these RF-A series?

It may not be entirely accurate to call RSS outdated, RSS exists because older and smaller ovens did/do not have the thermal output to produce a RTS process window that worked as they would struggle to deliver a ▲T across a board leaving you with all sorts of process issues. Fixing the issues by altering the ramp rate or peak temp introduces other issues and potentially damages sensitive components that want to keep TAL (or sometimes lower) as short as possible. Ideally you want a firmware that lets you adjust everything to match your paste board and component requirements as best as possible.

[F1GWR]

Almost all studies (google is your friend) show that RTS gives more reliable results than RSS, even, as you said, if there may exist some drawbacks. Nothing's perfect. But all in all RTS is better. In addition my solder paste recommended profile is given only for RTS.
My initial question therefore is still relevant.

[SMTech]

These studies all assume proper reflow ovens with 7 to 14 heating zones along a 4-8m heated tunnel. When we have that kind of equipment we are free to adjust each zone and the conveyor speed to deliver the reflow curve we are targeting and validate that using multiple probes on a test PCB to ensure the PCB and it components are being heated evenly.

With a crappy hotbox from China you are far more constrained by what your oven is capable of, sure someone could attempt to write a RTS profile, but the oven itself may have some inertia that means it more easily achieves something resembling RSS. More likely, it achieves something way off target for either, at least judging by some experiences voiced on this forum for the product type.

[48X24X48X]

When I got their oven many months back, I did mentioned this to them. Coming from Puhui's oven before that, which implements RTS instead of RSS, I was skeptical at first. Puhui's curve were tightly controlled but the random cold spots is just too random. After tweaking the default lead free profile given on the iTech/Zhengbang's oven, it does perform way much better and consistent than Puhui's RTS. This boils down to the better quality build of the oven itself with less leaking on the corners and joints. This is from my limited experience of only using small IR based oven.

[Kean]

After tweaking the default lead free profile given on the iTech/Zhengbang's oven, it does perform way much better and consistent than Puhui's RTS.

Same experience.  I'll probably be buying a second RF-A350 (or 250).

[JimboJack]

Hi can anyone give their reflow parameters for the  RF-A350 oven via the serial interface , or just use the default profile 4 etc ?


thks

J.

[GarbageIdeas]

We purchased one of these and are having a very hard time getting it to solder our boards. Itech themselves says that soldering this board isn’t possible in an rfa350 and the only support they can provide is to sell me a conveyor oven. We’re using AIM REL61 solder to get a little bit lower melt temp compared to SAC305. Normally we can’t get the paste to melt or we end up with scorched components as seen in the attached picture. Does this seem reasonable?

[m98]

We purchased one of these and are having a very hard time getting it to solder our boards. Itech themselves says that soldering this board isn’t possible in an rfa350 and the only support they can provide is to sell me a conveyor oven. We’re using AIM REL61 solder to get a little bit lower melt temp compared to SAC305. Normally we can’t get the paste to melt or we end up with scorched components as seen in the attached picture. Does this seem reasonable?

Have you tried using a more conventional and fresh SAC305 paste? What profile are you using? I have the RF-A250, and I have soldered much more demanding boards than that just fine.

[SMTech]

We purchased one of these and are having a very hard time getting it to solder our boards. Itech themselves says that soldering this board isn’t possible in an rfa350 and the only support they can provide is to sell me a conveyor oven. We’re using AIM REL61 solder to get a little bit lower melt temp compared to SAC305. Normally we can’t get the paste to melt or we end up with scorched components as seen in the attached picture. Does this seem reasonable?

Have you tried using a more conventional and fresh SAC305 paste? What profile are you using? I have the RF-A250, and I have soldered much more demanding boards than that just fine.

Hard to establish from purely a picture, maybe you got lucky.
 However these ovens from  use near IR heating rods. IR reflow has a bad name because it generates shadowing and uneven heating, and using this heating element type only makes that worse. You would get better performance using FIR heating panels but even so, plastic devices, and multiple electrolytics are not a great idea as these things absorb heat differently and cast shadows.
Its also quite difficult to get a nice relationship between the heat output of your lamps and the reflow profile measured on your board.

We used to run FIR 4&5 zone conveyor ovens, they could handle standard components with ease and the move to lead free was trivial. Reflow compatible connectors, push button switches, electrolytics next to each other, thick boards or larger inductors all an absolute no. (At least while sticking to anything resembling a compliant reflow time or TAL and you have to keep in mind everything else on the board).

[wraper]

Ovens like this have terrible heating uniformity and no bottom heating at all. Longer soak phase somewhat mitigates that as it allows heat to distribute over the board. RTS simply won't work on such HW unless you reflow a tiny board like 3x3 cm. And even then it will likely scorch components in plastic packaging.

[m98]

This is a convection oven, not pure IR heating like many others. Heat uniformity is actually pretty good.

[wraper]

This is a convection oven, not pure IR heating like many others. Heat uniformity is actually pretty good.

Convection, my ass. With IR heaters located a few cm above the PCB, fan is more like a gimmick rather than something that provides decent heat distribution. It's only somewhat effective during soak phase when heating power is turned down, not during rapid heating. Try actually putting multiple thermocouples inside such an oven. In center vs edges, and especially corners. And watch how even is temperature distribution during temperature ramp.
I once had a similar oven, with a fan. I made it somewhat useable by placing metal shields below the heaters to prevent direct IR heating in areas closest to the heater.

[Kean]

This is a convection oven, not pure IR heating like many others. Heat uniformity is actually pretty good.

Agreed.

Speaking from direct experience, these do have much better uniformity than the cheap Puhui T-962 and similar units.  I specifically got this oven on a recommendation for better uniformity, as my T962A would scorch parts in some locations, and yet leave other PCBs in a panel with unmolten solder paste.

It is certainly not perfect, and I'd love to replace it with a vapour phase unit.  I don't really have the space, power supply, or budget for a multi-zone conveyor oven.  I outsource bigger jobs, or my clients organise the manufacturing.

I also agree with SMTech that you will not easily be able to match a recommended reflow profile, but I think it should be able to do better than shown in that photo.

I don't normally have many electrolytics, large inductors, or large connectors in my designs, and those likely would need some extra soak time.  The few designs that I have with larger connectors (like test jig controllers), have typically used through hole parts.  I do use lots of smaller SMD connectors like JST PH/SH, Molex Picoblade, FPC, etc and I rarely have scorched parts.  I believe that the few times I have seen scorched plastic parts is when I run boards through immediately after a previous batch without allowing the oven to cool down sufficiently.

With respect to the PCB by GarbageIdeas, assuming it is a single sided load then I would likely use a hotplate for prototype assembly of a design like this.  With a hotplate you will need to carefully remove the PCB while still hot, as the plate will have too much thermal capacity to cool in a reasonable time.  Based on the business name, I don't think you are building mission critical parts.

You might want to consider converting the RF-A350 to another controller like the Reflowmaster Pro or Controleo3, as the default controller does provide very limited settings.  I've not tried this myself, so I don't know how much it would help.  The HQZB (Huaqi Zhengbang) HLH PC software is also pretty bad unfortunately, and I've been meaning to try reverse engineering the protocol.

I am generally happy with my RF-A350 for prototyping and small scale production, and have likely put hundreds of panels (a thousand plus boards) through it, mostly quite small but some as big as 250x250mm.  I've not recently done any profiling, but when I first got it I did run some profiling tests logging data from the RS232 port as well as using a separate 4-zone thermocouple logger, and was able to find some acceptable settings for my typical boards and preferred paste.

[Kean]

This is a convection oven, not pure IR heating like many others. Heat uniformity is actually pretty good.

Convection, my ass. With IR heaters located a few cm above the PCB, fan is more like a gimmick rather than something that provides decent heat distribution. It's only somewhat effective during soak phase when heating power is turned down, not during rapid heating. Try actually putting multiple thermocouples inside such an oven. In center vs edges, and especially corners. And watch how even is temperature distribution during temperature ramp.
I once had a similar oven, with a fan. I made it somewhat useable by placing metal shields below the heaters to prevent direct IR heating in areas closest to the heater.

You are right that it has the IR heaters not far above the PCB, but the air circulation has been reported to be quite effective by a number of users.  The heat reflective mat on the base of the drawer also seems to help.  Of course it is a compromise compared to bigger more expensive ovens, and it will not be ideal for some designs.

As mentioned above, it suits my current needs.  I do a lot of smaller PCBs like 30x30mm to 100x50mm in small panels, but occasionally I do bigger PCBs.  I try to avoid large thermal mass SMD parts, but I still use GPS modules with ceramic antennas (18x18mm), small inductors (e.g. SRR0745A or smaller), and various radio modules (ESP8266/ESP32/nRF52) with no issues.  Most issues that I have are with paste stencils and my "pre-loved" PnP equipment.

I am interested to know what oven you use now if you are happy to share.