Multimeter stopped working: Greenlee DM-45

[Selleli]

Hi.
My Greenlee DM-45 multimeter is only a few months old. It used to work fine until today. When I turn it on, the display remains off. When doing a continuity test (to check whether at least the buzzer still works) showed that even the buzzer doesn't sound anymore.
I checked the PCB to see whether anything would be amiss. But I couldn't find any problem (but I am beginner). Here is a picture of the PCB for everybody who is interested (see attachment "Full View small.jpg").

The only thing which caught my attention is the location marked at R48: There is no component mounted at that position (or it fell off). You can get a closer look to that specific location on the PCB in attachment "Full Zoom small.jpg".

Does anybody have the schematics of this PCB? Or does anybody have the same device and could post a close up picture of his/her PCB as comparison? 
Thanks.

[retiredcaps]

Some suggestions on fixing this.

1. Did you try a new fresh battery?
2. If you have a variable DC power supply, how much current does the DM-45 draw?
3. Did you measure something high voltage the last time it worked?  Or what did you measure the last time it worked?
4. Do you another multimeter to use to take measurements of the broken DM-45?

[Muttley Snickers]

Hi.
My Greenlee DM-45 multimeter is only a few months old. It used to work fine until today.

If your meter is only a few months old what is wrong with just returning it to the supplier for a warranty replacement, this would be the first thing on my list before dicking around with stuff and risk voiding warranties or further support.

[Selleli]

@Retiredcaps:

• Yes, I have tried a fresh battery already. It didn't help.
• Yes, I have a bench power supply. That's a good idea to check the current the multimeter draws. I have checked now: It draws 0.00A. Sounds like it could be a problem on the power input side of the circuit.
• No, I never measured high voltages. But I haven't used the multimeter for a month or so. And when I opened the battery panel, I have noticed that the negative terminal of the battery looked corroded or leaked. And when trying to remove the battery, the terminal broke off the battery without much force. However, the battery cable looks ok, and the voltage of the new battery / bench power supply is reaching the PCB. I could measure 9.6V at several test points on the PCB. So the battery cable didn't take any damage.
• I only have a cheapie secondary multimeter. Like a 10$ unit or so. But I have ordered a UNI-T UT61E now, just in case I won't be able to fix the Greenlee anymore.

@Muttley Snickers:
You are actually right. It would be reasonable to request a repair under warranty. However, the unit was not expensive (around 50$ or so) and it's actually a welcome opportunity for me to make my first steps trying to fix broken circuits. So I am actually glad that I have a broken device. Just too bad that it happens to be my most important measuring tool...   

[retiredcaps]

It draws 0.00A.

I could measure 9.6V at several test points on the PCB.

5. With the above two measurements, do you see any clock signals on Y1 (oscillator)?

6. Wrt to R48, it may be unpopulated by default?  RX2 is also unpopulated.

7. BTW, these are excellent pictures.  So good, I can tell the soldering is crappy.  I can tell the following have been hand soldered in: RX1, R11, R12, RX3, and R38.

8. Does Q8 look like there is a hole in it?  Or is that leftover flux residue on top of Q8?

[Selleli]

Thanks for your response, Retiredcaps! Much appreciated.
However, I think I might not be able to solve this one. Here is my follow up:

5. As I don't own an Oscilloscope (yet), I don't know how else I could measure a clock signal. Using my cheapie multimeter I switched it to 2000mA range and was able to measure that the voltage across the oscillator (Y1) is around 20mV. But not steady. I assume this is due to the fact that the measuring multimeter is trying to determine an average voltage over time, and the voltage is permanently varying. I guess with the limited measurement tools I have, the only conclusion I can draw is, that the oscillator is probably working fine. But that was a good trace, never the less.

6. You are right: I haven't noticed that before: RX2 is also unpopulated. And LED1 and LED2 as well. Hmmm. Seems to be nothing out of the norm then...

7. Thanks for the picture quality compliment! Took me quite some effort to build up the tripod and all camera gear for those shots. Yes, indeed the components you mention seem to be hand soldered in. I have tried re-soldering Rx1 using my brand new heat gun. It looks better now. But it didn't fix the multimeter.

8. Q8 looked almost like it had an edge broken off on the picture, indeed. But I have checked it again, and it was indeed just some flux residue. After cleaning it up, it looks totally fine.

I should add that yesterday I was sometimes able to get a faint reading on the display of the multimeter when switching through the different modes. But the reading was weird and didn't make any sense. And it always faded after a second or two. And today I am not able to produce the faint readings on the display anymore. Even yesterday I was initially not able to produce these faint readings. I seemed like a random act of kindness of the multimeter. Like throwing me a cookie. 

[mzacharias]

Does the main IC or any other component seem unusually warm after power on for say one minute? Multimeters of this type are generally pretty low current drain - so anything very warm is probably leaky or shorted.

[Selleli]

@Mzacharias: Thanks for the lead. That sounds worth being checked! So I have done that now. And even after several minutes, no component gets warm enough that I could notice any difference with my bare hand. And the fact that the multimeter under test is drawing 0.00A when being turned on, also seems to indicate that it might be something other than a shorted component or shorted circuit.

I have now also opened up the display unit but haven't found anything which looks alarming. And the display unit is receiving around 400mV on its two power input leads.

I guess with the limited measuring tools, absent schematics and my still limited skills, I am coming to a conclusion that this problem is perhaps too tricky for me to solve.  But that's fine. Thanks everyone for having provided helpful advice. I have learnt a bit more already. 

[ModemHead]

I may be taking your report too literally, but "0.00A" is not enough resolution to determine whether the meter is drawing any current or not.  A 9V-powered DMM would probably draw 1 to 5 milli-amps normally.

Use the milli-amp range on your working DMM to measure.  If the faulty DMM starts drawing a nominal amount of current when turned on, then perhaps you're right, it may be a tricky problem for you.  But if it is truly drawing zero current, it could be as simple as a broken/corroded PCB trace, perhaps caused by the leaky battery.

[Selleli]

@ModemHead: Thanks for the idea! That's a good point. Previously I really didn't have more resolution than just 0.00A, because my bench power supply doesn't display any higher resolution for the current. But you are perhaps right that a DMM would perhaps draw considerably less than 0.01A and hence it might never show any reading on the ammeter of the power supply.
I have used my working DMM in mA mode to measure the current drawn by the DMM under test when it's turned on, and in different modes. And the readings on the working DMM ranged from 0.4mA to 1.5mA. So the DMM under test is really drawing some power when it's turned on. Mostly around 0.4mA. Which seems almost too little. But it's clearly not zero. Consistently. Which makes me believe that it shouldn't be a broken/corroded PCB trace which is the problem in this case.
Was worth checking though.

[Davy]

If the meter isn't drawing any current then nothing could possibly get warm as I can see.... would it be worth mentioning the soldered connections to the impregnated chip on the PCB, pin 3 on the first photo?

I don't know, maybe lead free solder, shadowing.... I've repaired faults, not meters though where the soldering just looks like that around pins 1~3 and the connections been opened circuit.... it's as though the connections are water affected.  Just thought I would mention.

If there's been any leakage, I'd be checking the plated through holes either side of the board, I repaired a Fluke two weeks ago that was intermittent on AC volts, due to a connection on the rotary switch.

Dave

 

[Davy]

How does it compare to these photos?
https://www.eevblog.com/forum/testgear/dm-45-review/

Dave

[LA7SJA]

Can you try to measure the resistance from the 10Amp connector and the Com connector? The 10 amp shunt looks like it has a bad solder joint. If you overload with current the fuse breaks but if the shunt breaks/disconnects the "Chip" can die from overvoltage when you try to measure current. If you have a case of overvoltage you should see something "funny" under the range switch.

Johan-Fredrik

[Selleli]

@LA7SJA: I see which solder joint you mean: The port of the 10A connector looks indeed quite bad. However, I don't think I have used the 10A port before, as I am only measuring low currents. I have now measured the resistance between the 10A connector and the Com connector: It's 0 Ohms. The resistance between the low amps connector and the Com connector is huge: It's above 1M Ohms.

I have already removed the range knob/switch and everything looked normal underneath it. 

[LA7SJA]

Can you measure voltage on pin 4 U1 and the voltages on U3 (can you identify it?) all pins. Do you measure any voltages at all on U1?

Johan-Fredrik

[kcbrown]

How does it compare to these photos?
https://www.eevblog.com/forum/testgear/dm-45-review/

It looks like RX3 is unpopulated in the review meter, while R48 is -- the opposite of what we see in the photo from this thread.   Notably, it appears there's been some rework of that area in the meter in this thread, since it looks like there's flux residue in that area, as if someone moved the resistor from R48 to RX3 or something.

OP: what kind of chip is U3?  I second the notion of examining the voltages on it.  Get the datasheet for it and see what it is.  If it's a power control IC or something, then it might be at issue here.   You won't know what to expect to see on its pins or how it might affect the meter without knowing what it is, so the first order of business is to identify it and get the datasheet.

[Gregg]

RX3 and R48 appear to be padding for VR6 to enable hands on calibration during final assembly and in this meter somebody decided that only RX3 to make VR6 range work.  The flux that thee calibration person left on the board may have gotten into VR6 making a poor connection.  It may be worth trying to clean off the flux and use some quality contact cleaner on VR6.  Also it probably wouldn’t hurt to re-solder all of the connections that look hand soldered.  Another item of curiosity is the hand soldered via next to RX3; it seems strange that it has no trace on the pictured side of the board if it is only a two layer board.  If it is more than two layers, whoever hand soldered it may have compromised the intended connection.

[JFJ]

Does anybody have the schematics of this PCB? Or does anybody have the same device and could post a close up picture of his/her PCB as comparison?

You might increase your chances of finding a schematic (or pictures of the PCB) if you can discover other brand names and model numbers for your meter. The model number used by the meter's actual manufacturer (all-sun) is EM6186:


(from http://www.all-sun.com/EN/d.aspx?pht=1378)

Alternatively, if you can identify the main controller chip (the one sealed under the black epoxy) then its datasheet could be almost as useful as a meter schematic. That's because the meter manufacturers who use single chip DMM controllers, appear to stick to the circuit diagrams provided in the chip's application notes.

It may be possible to deduce which controller chip has be deployed, in your meter, from the connections to surrounding QFP-100 solder pads. Comparing which pads connect to your meter's LCD module and crystal with the pinout for DMM chips in QFP-100 packages, can quickly eliminate incompatible controllers. For example, the popular FS9721-LP3 DMM chip can be eliminated, because pins 80 and 81 are common connections to the LCD module in the datasheet, but are not connected on your PCB.



Tracking down datasheets for DMM controllers can be time consuming, but you did say "... it's actually a welcome opportunity for me to make my first steps trying to fix broken circuits". To get you started, here are a couple of links to DMM chip manufacturer datasheets:

http://www.cyrustek.com.tw/product-1.htm
http://www.ic-fortune.com/eng/new_product3_3.asp

Perhaps, other forum members have additional links?

[ModemHead]

When I first saw this thread I happened to have the data  sheet for the Fortune Semi FS98O24 on my desk.  I noted that the crystal pins (69 & 70) match exactly, and the voltage divider and LCD lines seem to match up in a general way.  Doesn't seem like an appropriate chip for this model, but it could be from the same family?

[Selleli]

@JFJ:
Wow, thanks for your elaborate response! How did you know that my Greenlee DM-45 is the same model as the EM6186 by All-Sun? Just from your experience or is there any way how I could have found out by myself?

I understand: You are saying that by identifying the DMM controller, I could get helpful schematics from the DMM controller manufacturer, which usually comes quite close to the final schematics of the DMM. So this would compensate for the lack of the original schematics of the DMM manufacturer. Clever!
I will try to use this knock-out scheme tomorrow to narrow down the potential candidate list of DMM controllers. And then see what I am left with.

@ModemHead:
Yes, now I have checked the datasheet http://www.ic-fortune.com/upload/Download/FS98O24-DS-16_EN.pdf in regard to pins 69 and 70. And they seem to match, yes. I will check more closely, and also consider other controllers of the same family later. Thanks.

[retiredcaps]

How did you know that my Greenlee DM-45 is the same model as the EM6186 by All-Sun?

All-sun is silk screened on the pcb in your first photo.

In addition, most tool companies like Greenlee, Southwire, Klein, Extech don't make their own multimeters.  They just rebrand meters made by someone else in their own color schemes.

[JFJ]

How did you know that my Greenlee DM-45 is the same model as the EM6186 by All-Sun?

The all-sun logo is printed on your meter's PCB (Re: top-right corner of your first picture), and only the EM6186 has the same specifications as the DM-45.

[JFJ]

All-sun is silk screened on the pcb in your first photo.

Snap!