new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004

[Robaroni]

The world has changed since 2017, my next scope will be from China.

[Hydron]

Welp, just had another channel go bad in mine. Same symptoms as before, huge offset has developed in part of the PGA range. There is now finally a reply on the TI forum expanding on the datasheet change, albeit in a somewhat confusing fashion: https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1569255/lmh6518-offset-problem-with-lmh6518

@pdenisowski - does the "new" RTB2x range have this same LMH6518 PGA inside with the same front end design (i.e. with an output range from the PGA that can exceed 1.7V in steady state conditions)? I think this issue is bad enough that the product shouldn't be on the market until fixed if so, as it's a ticking time bomb that will ruin instruments through no fault of the user. To be clear I'm not saying this is R&S's fault - the original datasheet did NOT have the max output note in the abs. max conditions section, but it is now a known fatal issue with the product.

I'd also like to know what is going to be done about those of us with existing units, especially outside of warranty - the official repair costs are frankly punative for those of us who bought them for hobby use as originally encouraged by R&S here on the forum (scaling the price by BW license is totally unreasonable - repair should be on a cost recovery basis, not a way to extract maximum money from a helpless customer).

[exe]

in a somewhat confusing fashion: https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1569255/lmh6518-offset-problem-with-lmh6518

I broke my brain trying to parse that. I feel like they are caught in action, and yet they still pretend to be innocent. Probably they spent many hours to write that.

[Hydron]

I THINK it's trying to say that the issue can occur at any time when the gain is set such that the output is >1700mV, even if the input is <400mV.

And then an ass covering type got involved too and mangled it further.

The fact remains that a major piece of design information was missing for years from the datasheet, and was something so fundamental that it both should have been caught in testing, and will impact many users of the part.

I gotta say that the output voltage limit being the critical thing doesn't really line up with the block diagram and symptoms (offset goes up but only for the low gain region, which the block diagram suggests is right at the input), but maybe the way it's presented is quite different to the actual construction.

I'll have a look at what the actual output range is when I open up my scope to fix it (assuming R&S don't come back soon with a reasonable solution that isn't paying more than the scope is worth for a board replacement like they quoted last time).

[voltsandjolts]

...when I open up my scope to fix it...

Photos please!

[nctnico]

in a somewhat confusing fashion: https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1569255/lmh6518-offset-problem-with-lmh6518

I broke my brain trying to parse that. I feel like they are caught in action, and yet they still pretend to be innocent. Probably they spent many hours to write that.

The message is quite simple: maximum DC output voltage is 1700mV (pp). The problem is not the input stage, but the output stage of the chip.

Anyway, IMHO R&S should fix scopes for free if they have this problem. It sounds like a relatively easy rework job.

[Hydron]

Confusion comes from the failure previously being discussed as an input voltage issue, as noted even in TI's post:

"I did communicate and discuss an input voltage limit with multiple customers & on e2e, such as a steady-state or DC version of the existing "Differential input signal voltage ±1 V" spec.  My work resulted in the working guidance of 400mV, which you mention and various engineers have now become aware."

I was just noting that it is surprising that it's an output thing, given the failure mode and datasheet diagram - again it could be much more complicated internally than is presented in the datasheet however, and they don't really explain what is really happening, possibly because it could give customers who have been burnt by this fuckup more legal ammunition.

Really fucking poor from TI all around I have to say, firstly missing a major product design/documentation issue, then with this confusing mess where they change the datasheet without proper explanation and need a bunch of prodding to get even a vague answer.

[Darkover]

Anyway, IMHO R&S should fix scopes for free if they have this problem. It sounds like a relatively easy rework job.

R&S fix it for free if you are in warranty. There solution is to replace the complete mainboard of the scope. Of course that is
not an acceptable solution if you have to pay for it. 

At the moment my scope is open and I try to replace the chip. But that is MUCH harder than I expected and I am experienced in this kind of work. I think 90% of this board is made of copper. For example a normal hotplate will not work. It sucks heat like a frozen hell. I will post some picture when I am ready and when it is working again. Or perhaps when it looks like a sliced toast bread. But that takes time, because I need some mediation and a tibetian swinging bowl for this work.

But that is not the biggest problem. If we replace the chip and everything it working again, it is easy to destroy it again when doing the same measurement. R&S has to provide a different firmware that use different calibration/voltage level when working in DC-Mode.

I think the most interesting question at this time, what is with there new and shiny RTB2? They should know about the problem, because they had to repair some/many boards in the past. So did they change the hardware? Did the V3.0 solved the problem? Or is it still there? Can I buy a new and shiny RTB2 and send it back for repairing every three weeks in the first two years?

Olaf

[Hydron]

At the moment my scope is open and I try to replace the chip. But that is MUCH harder than I expected and I am experienced in this kind of work. I think 90% of this board is made of copper. For example a normal hotplate will not work. It sucks heat like a frozen hell. I will post some picture when I am ready and when it is working again. Or perhaps when it looks like a sliced toast bread. But that takes time, because I need some mediation and a tibetian swinging bowl for this work.

But that is not the biggest problem. If we replace the chip and everything it working again, it is easy to destroy it again when doing the same measurement. R&S has to provide a different firmware that use different calibration/voltage level when working in DC-Mode.

I think the most interesting question at this time, what is with there new and shiny RTB2? They should know about the problem, because they had to repair some/many boards in the past. So did they change the hardware? Did the V3.0 solved the problem? Or is it still there? Can I buy a new and shiny RTB2 and send it back for repairing every three weeks in the first two years?

Yes it was difficult - needed a LOT of heat before it would reflow, and I had to go back in and touch up a couple of pads. I am hoping it will be easier this time around as I have a hotplate for pre-heat and a stereo microscope for checking my work. Might also use low-melt solder paste (definitely NOT lead-free!), though that won't help getting it off in the first place.

As for RTB2x and FW for existing units, this is exactly why I @'d the R&S rep on the forum - this is not an issue that can be ignored given the consequence is a dead scope.

When I fix mine I will check the DC output level of the PGA vs attenutation and offset so I know what exact situations put it at risk.

A firmware workaround (NOT fix - that would need a HW change) would either need to allow much greater offset during self-cal (like early <= v2.000 FW does, probably at the cost of offset adjust range) and/or adjust gain to avoid high DC levels on the output (probably increasing noise, and may not be possible at all for the 5V/div range).

[Darkover]

Yes it was difficult - needed a LOT of heat before it would reflow, and I had to go back in and touch up a couple of pads. I am hoping it will be easier this time around as I have a hotplate for pre-heat and a stereo microscope for checking my work. Might also use low-melt solder paste (definitely NOT lead-free!), though that won't help getting it off in the first place.

I am thinking in the same direction! The absolute minimum is leadsolder for 183°C. I also have some solder paste with wismut that has a lower melting point. However that solder feels like hot pea soup. I don't like to use it. But I have to investigate in it.

Olaf

p.s: I try to share some picture from the current working state. 

[Darkover]

I like to say the picture only showed the preheating! I had a second heat gun on top. But it was not strong enought.

Let me share a picture from the actual state of work! 

The board is still okay, but of course the original LHM6518 has no offset problem anymore.

And I would never ever try something like this without a 3D microscope. It is an very interesting experience to look at 3D inside of the LHM6518 when the plastik housing falls apart, but the exposed pad is still there.

The next, but now untested idea is to use a Weller pyropen for heating on top. But I feel a little timidly now.  The pyropen has much more energy and can easy roast the board.

Olaf

[Hydron]

Woah, I had nowhere near that level of damage getting mine off! Just did it with a 800W hot air gun and a lot of patience (I looked for pics, couldn't find any, it was 5 years ago!).

[goaty]

OMG this looks ruined.
I think of selling my RTB before it develops the same problems.
But then this is kinda unfair to the potential buyer.

[Robaroni]

I like to say the picture only showed the preheating! I had a second heat gun on top. But it was not strong enought.

Let me share a picture from the actual state of work! 

The board is still okay, but of course the original LHM6518 has no offset problem anymore.

And I would never ever try something like this without a 3D microscope. It is an very interesting experience to look at 3D inside of the LHM6518 when the plastik housing falls apart, but the exposed pad is still there.

The next, but now untested idea is to use a Weller pyropen for heating on top. But I feel a little timidly now.  The pyropen has much more energy and can easy roast the board.

Olaf

hi Olaf,
Do you, or does anyone else, have picture of the chip before heating it?

[gdr771]

Anyone know what RTB2004 setting(s) and/or signal input value cause the LMH6518 to go into saturation (>=1700mVp-p)?

[goaty]

hi Olaf,
Do you, or does anyone else, have picture of the chip before heating it?
[/quote]

Shariar has pictures at about 8:30
https://youtu.be/YokF2_EbfIk?t=517

[Darkover]

Do you, or does anyone else, have picture of the chip before heating it?

What did you expext from a picture? It is a normal 16pin QFN with EP. There is nothing special with this chip. I solderd chips like this many times! The problem in this case is that this chip has a normal power consumption of 1W and for this reason the PCB is full of copper. R&S delivered us a realy good high quality PCB for our money! The only problem they did not use a socket. 

Olaf

[Hydron]

hi Olaf,
Do you, or does anyone else, have picture of the chip before heating it?

Shariar has pictures at about 8:30
https://youtu.be/YokF2_EbfIk?t=517

[nctnico]

I like to say the picture only showed the preheating! I had a second heat gun on top. But it was not strong enought.

I use a hotplate for these kind of jobs. That works much better compared to a heat gun as a hotplate supplies heat to the board over a much larger eara which is what you need when dealing with a multilayer board. The board will get rid of heat quickly if you heat only a small area (and you need to make that small area very hot; much hotter than is safe for the components).

BTW, an interesting problem with using heat guns is that having it too close, means it doesn't heat very well because the airflow gets obstructed.

[Darkover]

I use a hotplate for these kind of jobs.

As I said before, using a huge hotplate was my first idea and try, too. Putting the PCB on a 200°C hotplate will only give you 100degree on the topside, because the bottom is populated, too. (Measured by infrared thermometer) But I agree with you in one point. I thought perhaps trying to increase the distance to the hot gun. However, you have seen that the bottom was at 160°C and I measured 150°C on the topside before I started to used a second hot gun on topside.
Perhaps using a infrared heater would be the best, or removing the parts from the bottom side to the PCB can be flat on a hotplate. However many of the part are the shielding connector with good ground connection, too. I thought perhaps I need a bed of sand on the hotplate, but until now it is now tested.

Olaf

[Hydron]

I'm really starting to wonder how on earth I fixed mine the first time without even a hotplate helping or a microscope available.

Might be a few days before I have a go - gotta find time, make some space for the soldering setup and be ready to be without the scope for a bit.

[nctnico]

I use a hotplate for these kind of jobs.

As I said before, using a huge hotplate was my first idea and try, too. Putting the PCB on a 200°C hotplate will only give you 100degree on the topside, because the bottom is populated, too. (Measured by infrared thermometer)

I know. But having the PCB heated to 100 degrees already helps a lot to get 'stubborn' parts removed using hot air. Typically I have my hotplate set to 250 degrees and start rework when the PCB has reached 100 degrees. I use hot air set to 330 degrees or 340 degrees at most.

[Darkover]

Typically I have my hotplate set to 250 degrees and start rework when the PCB has reached 100 degrees. I use hot air set to 330 degrees or 340 degrees at most.

Whow! Is 250°C not a little bit to much? I would worry for the bottom parts in this case. I startet at 150°C and use hot air on top wht 370°C.

Olaf

[Hydron]

I suspect it's a bit different using a hotplate vs a heatgun - the hotplate needs to be significantly hotter than you want the board to get, at least until things reach steady state. Will see what I need when I do mine.

[Darkover]

It feels a little, this is a coocking group now. 

Olaf