I bought a development board from Mouser last week and when the board arrived it was defective. I called Mouser customer service expecting them to give me an RMA number so I could send it back for a replacement, but no, they said I’d have to send them a detailed explanation of the issue and they’d contact ST to determine what they wanted to do.
If this were a $15 Nucleo board, I wouldn’t bother—I’d just toss it on the junk pile and move on. But this is a $330 board that I need for a project. Now I have to wait indefinitely while Mouser works this issue with ST. I probably won’t have a working board for weeks.
Is this level of “customer service” common in the industry? I suppose I’m spoiled by Amazon’s no questions asked return policy.
I imagine that they want to verify both that your problem is actually a defect and that a replacement will solve the issue. Amazons 'dumb' return policy is handy when you run into stupid or slippery sellers, but it doesn't do you much good as far as obtaining a good or correct product. If you called Mouser and a human answered the phone in a reasonable period of time, that puts them in the top 10% for CS these days.
My experience with Mouser is a bit dated, but it has been quite positive. However, the amount in question was never more than $50, which is trivial. One notable example was when I got on a spammers email list. No money per se was involved. Despite many attempts to "unsubscribe," it continued. Then I noticed Mouser advertised on the spammer's site. In desperation, I wrote Mouser and got a personal phone call back. After explaining, the person said she would look into it. All emails from that spammer stopped immediately. That was beyond what I expected and made a lasting impression.
I bought a development board from Mouser last week and when the board arrived it was defective. I called Mouser customer service expecting them to give me an RMA number so I could send it back for a replacement, but no, they said I’d have to send them a detailed explanation of the issue and they’d contact ST to determine what they wanted to do.
If this were a $15 Nucleo board, I wouldn’t bother—I’d just toss it on the junk pile and move on. But this is a $330 board that I need for a project. Now I have to wait indefinitely while Mouser works this issue with ST. I probably won’t have a working board for weeks.
Is this level of “customer service” common in the industry? I suppose I’m spoiled by Amazon’s no questions asked return policy.
Yes you are spoiled...
I have worked through several customer returns and not one of them went as "no questions asked". I would say that, if you can provide the most details to Mouser or ST, the processing of this return will be faster. Even if the kit is "Dead on Arrival", details such as voltages present/absent, cables replaced/tested, etc. Will go a long way to speed up the process.
Keep in mind that, if this is the last stocked kit, there is always a chance that a full replacement will take a while to be shipped and sometimes is advantageous to try your best to identify the root cause of the issue.
(Obviously that I have no idea what is the extent you have gone with Mouser).
I do understand Mouser is critical in taking returns and offer replacements. I know for myself of all my friends/family and acquaintances there are individuals who would jump quickly to the conclusion something is defective. You might ask yourself why should Mouser have blind faith in the judgement of each and everyone.
So.. what was the problem and how did you determine it?
Their return policy is not very clear, but, if you have a good credit card I would consider buying another one and assume that you'll be able to return the first order for a full refund (if they have more in stock).
https://www.mouser.com/help/orders-shipping/return-policy/
Sure, Mouser perhaps shouldn't have blind faith in the "judgement of each and everyone", but how far should one have to go to get satisfaction from a company after paying them a considerable amount of money for a product that doesn't work?
In my case, I spent several hours probing the board looking at voltages, checking continuity from connectors to MCU pins, and I even looked at the Gerbers to see if there were any potential shorts or signals connected to nets they shouldn't be.
Yes, I could have gone further and removed and replaced components to try to troubleshoot the issue, but ultimately didn't for two reasons. Reason one: this board has lots of tiny surface mount parts packed tightly together. Reason two: why should I? This is a brand new board I paid several hundred dollars for, so why should I be expected to debug it on a hardware level, particularly when Mouser could then claim I screwed the board up with my debugging.
While I was typing this reply, someone asked what the problem was. Here's a description. This board is a medium power motor control board that is designed to drive three-phase motors up to 75v and 35A. It has an interface for Hall and quadrature encoders. In the default configuration, the three encoder inputs have 10K pull-ups to Vdd. I connected a motor with Hall position sensors to the board and tested the encoder inputs with some simple code that reads the three inputs and displays the Hall state over a serial port. Rotating the motor's shaft by hand I was seeing only even numbers in the range 0-6. I should have seen even and odd numbers in the range 1-6 (0 and 7 are forbidden Hall states). This implied Hall input #1 was stuck at "0". I know this motor's Hall encoder works because I immediately tested it on another board and saw the expected outputs on its Hall encoder when I turned the motor shaft.
As another test, I disconnected the motor from the board. Since the encoder inputs are pulled up to Vdd, I was expecting to see "1" on each line ("7" when the three where concatenated). I saw "6", again indicating Hall input #1 was stuck low.
The Hall encoder circuit is simple. There's a 10K resistor in-line to limit current, a 10K pull-up to Vdd (3.3v), a 1nF cap to ground to filter noise, and a Schottky diode to Vdd to clamp input voltages to the Vdd rail. Each of these three circuits (one for each encoder input) then goes directly to an MCU pin, which in the input state is hi-Z. When I put 3.3v on the H2 and H3 inputs (the working ones), I saw 3.3v on both sides of the 3.3K resistor in-line from the input to the MCU. When I put 3.3v on H1, I saw 3.3v on the connector side of H1 and 1.25v on the MCU side of the resistor. Probing directly at the MCU pin, I saw the same 1.25v. The MCU's threshold for a logic high is 1.88v, which would explain why I'm always seeing "0" on that input. Something is pulling down that input, but without removing components from the board, it would be difficult to tell what it is.
According to the schematics, there's nothing else connected to this MCU input pin, and I verified this by looking at the Gerbers. I also verified that all three of the MCU ports associated with the encoder inputs are in input mode with the internal pull-ups/pull-downs disabled.
What matters now is how you're going to present this to Mouser in a convincing way.
Just telling them "It's defective" isn't convincing.
Sal, as far as I can tell that explanation makes sense but I am not surprised at all that Mouser deferred this to the manufacturer. You might be connected to an FAE or dev engineer that is familiar with the kit. Mouser couldn't possibly give judgement of pass/fail on such complex description.
The Hall encoder circuit is simple. There's a 10K resistor in-line to limit current, a 10K pull-up to Vdd (3.3v), a 1nF cap to ground to filter noise, and a Schottky diode to Vdd to clamp input voltages to the Vdd rail. Each of these three circuits (one for each encoder input) then goes directly to an MCU pin, which in the input state is hi-Z. When I put 3.3v on the H2 and H3 inputs (the working ones), I saw 3.3v on both sides of the 3.3K resistor in-line from the input to the MCU. When I put 3.3v on H1, I saw 3.3v on the connector side of H1 and 1.25v on the MCU side of the resistor. Probing directly at the MCU pin, I saw the same 1.25v. The MCU's threshold for a logic high is 1.88v, which would explain why I'm always seeing "0" on that input. Something is pulling down that input, but without removing components from the board, it would be difficult to tell what it is.
According to the schematics, there's nothing else connected to this MCU input pin, and I verified this by looking at the Gerbers. I also verified that all three of the MCU ports associated with the encoder inputs are in input mode with the internal pull-ups/pull-downs disabled.
If you check the resistance to ground with the power off, is there a major difference on those two inputs? Directly across the 1nF cap.
Whatever it is is a 0.6mA draw.. so not very much.
Nothing else is connected to those mcu inputs on the schematic right.
Maybe hold the thing in reset and see if injecting 3.3V has the same result. Although, with the pin set as input, should be confident already..
What matters now is how you're going to present this to Mouser in a convincing way.
Just telling them "It's defective" isn't convincing.
I've sent them a description of what I found, similar to the description above.
The Hall encoder circuit is simple. There's a 10K resistor in-line to limit current, a 10K pull-up to Vdd (3.3v), a 1nF cap to ground to filter noise, and a Schottky diode to Vdd to clamp input voltages to the Vdd rail. Each of these three circuits (one for each encoder input) then goes directly to an MCU pin, which in the input state is hi-Z. When I put 3.3v on the H2 and H3 inputs (the working ones), I saw 3.3v on both sides of the 3.3K resistor in-line from the input to the MCU. When I put 3.3v on H1, I saw 3.3v on the connector side of H1 and 1.25v on the MCU side of the resistor. Probing directly at the MCU pin, I saw the same 1.25v. The MCU's threshold for a logic high is 1.88v, which would explain why I'm always seeing "0" on that input. Something is pulling down that input, but without removing components from the board, it would be difficult to tell what it is.
According to the schematics, there's nothing else connected to this MCU input pin, and I verified this by looking at the Gerbers. I also verified that all three of the MCU ports associated with the encoder inputs are in input mode with the internal pull-ups/pull-downs disabled.
If you check the resistance to ground with the power off, is there a major difference on those two inputs? Directly across the 1nF cap.
Whatever it is is a 0.6mA draw.. so not very much.
Nothing else is connected to those mcu inputs on the schematic right.
Maybe hold the thing in reset and see if injecting 3.3V has the same result. Although, with the pin set as input, should be confident already..
With power off, resistance to ground is 13.25K on all three inputs.
Yes, there's nothing else connected to those three MCU inputs. I've tried injecting 3.3v with the MCU held in reset and the results are the same: 3.3v at the two "good" inputs and 1.25v at the "bad" input as measured directly at the MCU pins.
I bought a development board from Mouser last week and when the board arrived it was defective. I called Mouser customer service expecting them to give me an RMA number so I could send it back for a replacement, but no, they said I’d have to send them a detailed explanation of the issue and they’d contact ST to determine what they wanted to do.
If this were a $15 Nucleo board, I wouldn’t bother—I’d just toss it on the junk pile and move on. But this is a $330 board that I need for a project. Now I have to wait indefinitely while Mouser works this issue with ST. I probably won’t have a working board for weeks.
Is this level of “customer service” common in the industry? I suppose I’m spoiled by Amazon’s no questions asked return policy.
Why didn't you take this up with the manufacturer?
Companies and private customers have different e-shopping rights. In NL too, we can basically send anything back for up to 2 weeks within reasonable usage terms (e.g. just to see if they will suit you, so don't wear clothes to a party and then send them back)
I've heard stories about *several* 5k$ FPGA boards being sold to my university, that were then a few months down the line picked up for a project. They were all defective in one way or another. Contacted support, they confirmed the issue but 1) Low volume product, bad luck. 2) Apparently waited a few months for evaluation is too long. 3) Not private customer, so though luck.
On top of that, these are technical components.. so unless you have some kind of conversation with the manufacturer that can show your board is indeed defective, then it becomes very hard to proof you haven't read the manual incorrectly (which, by the expectation that manuals are *written* incorrectly, then has a 100% probability of occuring).
I bought a development board from Mouser last week and when the board arrived it was defective. I called Mouser customer service expecting them to give me an RMA number so I could send it back for a replacement, but no, they said I’d have to send them a detailed explanation of the issue and they’d contact ST to determine what they wanted to do.
If this were a $15 Nucleo board, I wouldn’t bother—I’d just toss it on the junk pile and move on. But this is a $330 board that I need for a project. Now I have to wait indefinitely while Mouser works this issue with ST. I probably won’t have a working board for weeks.
Is this level of “customer service” common in the industry? I suppose I’m spoiled by Amazon’s no questions asked return policy.
Why didn't you take this up with the manufacturer?
Because the company I bought the board from, Mouser, is likely to have much more clout with and access to the right people at ST than I do. If I don't get a satisfactory response from Mouser then I'll contact ST directly.
You wanna try Amazon CS then... anything more than a trivial swap/replacement issue, going a bit deeper, you'll soon lose the will to live!
You wanna try Amazon CS then... anything more than a trivial swap/replacement issue, going a bit deeper, you'll soon lose the will to live!
In over forty years of buying electronics, I've been very lucky and have had very, very few bad or DOA components and products. The vast majority of which were, fortunately, purchased from companies that accepted returns without much fuss. I'm still waiting for Mouser's reply, and hopefully they'll either have a simple fix that I can implement myself, or accept a return and send me a working replacement.
I've bought lots of stuff from Amazon, and I've only had to return one item--a book that got mangled in shipment. Amazon sent me a replacement and didn't even ask me to return the original book.
You wanna try Amazon CS then... anything more than a trivial swap/replacement issue, going a bit deeper, you'll soon lose the will to live!
In over forty years of buying electronics, I've been very lucky and have had very, very few bad or DOA components and products. The vast majority of which were, fortunately, purchased from companies that accepted returns without much fuss. I'm still waiting for Mouser's reply, and hopefully they'll either have a simple fix that I can implement myself, or accept a return and send me a working replacement.
I've bought lots of stuff from Amazon, and I've only had to return one item--a book that got mangled in shipment. Amazon sent me a replacement and didn't even ask me to return the original book.
Read my post again, carefully...
You wanna try Amazon CS then... anything more than a trivial swap/replacement issue, going a bit deeper, you'll soon lose the will to live!
When one receives a defective product or one that is not as advertised what else besides replacement or refund would you want?
That is, what do you mean by "going a bit deeper?" That could mean anything from instructions to consequential damages.
When one receives a defective product or one that is not as advertised what else besides replacement or refund would you want?
It depends on what it is and the nature of the problem. If a replacement will fix the issue, then that is what you want. But if you have an issue--say something like a firmware bug or design issue--then unit replacement is just going to be more of the same. So what you want is a manufacturer (not a reseller) with a responsive support team that will understand the issue at the very least. I would assume that a manufacturer like ST would want to actively support something like a development board. But nowadays, who knows.
While I agree about expectations for a manufacturer, my comment was in direct response to eti's comment about Amazon. To my knowledge, the only tangible it produces is money.
Now, if you happen to buy a Blue Origin rocket, you might want a little "deeper" customer service.
Just heard back from Mouser. ST wants to know what my "volume" is for this board. I fail to see how that's relevant to the issue at hand. Is ST going to tell me to pound sand because I will only ever buy one of these boards?
Maybe I should tell them I'm planning on at least 100,000 a year.
I suspect that response was from some functionary that didn't know what a development board was for. I would answer honestly, i.e., 1, and add that the board is for development of other devices. For such a low number, it might just authorize Mouser to send you another board gratis.