RCBS Chargemaster 1500 scale & powder dispenser

[TortureGarden]

Good morning all.

I have an RCBS Chargemaster 1500 combo electronic scale/powder dispenser that I have been using without issue for 15 years or more. I powered her up to do some reloading the other week and the LCD panel backlight & characters lit-up and then immediately, the characters faded-out, leaving just the backlight illuminated. I thought it may have been a one-off so, I tested again (and again...), with the same result.

I separated the scale from the dispenser and disassembled the scale. Nothing obvious observed that could be causing the issue, caps look okay. Reconnected to power and control panel and powered-on. Gently exerted pressure along the edges of the LCD and still no image.

Removed the LCD panel, donned the gloves and cleaned the elastomeric connectors on all sides and contact area on the LCD. They were slightly dirty. Reassembled and re-tested. Same issue.

Pulled out the multi-meter and did a few readings, not knowing what the various voltage on the board should be;

• Input - 9.19V
• LCD backlight - 2.85V
• LCD - 5.0V

I looked-up the microcontroller of the unit and it's a Renesas 8-Bit Single-Chip Microcomputer H8/38347.

Far from being an electronics guru, any help would be greatly appreciated.

Cheers, James.

[stj]

warm up the caps with a hair dryer and see if it works

[TortureGarden]

warm up the caps with a hair dryer and see if it works

Thanks mate. Will do in the morning and report back.

[TortureGarden]

Okay. So, heated the caps with a hot air gun set to 100°c and unfortunately the issue persists although, the characters did appear to appear for a slightly longer period of time on the LCD

[CapLeaker]

So it isn't a cap issue then. Would have kind of surprised me, because the display does come on, but then clunks out. ICs do have that kind of problem. So do dc to dc converters. See if a low power voltage is missing when it clunks out.

why do is some pics the caps look wet?

[TortureGarden]

So it isn't a cap issue then. Would have kind of surprised me, because the display does come on, but then clunks out. ICs do have that kind of problem. So do dc to dc converters. See if a low power voltage is missing when it clunks out

Hi mate. Thanks for the input. Being a complete newbie, where would I look on the PCB to check the voltages?

why do is some pics the caps look wet?

The board appears to be entirely covered with a gel-type coating; water-proofing or antistatic perhaps?

[CapLeaker]

Start with that 7806 6V regulator. See. These there is sufficient power getting to it and from it. There also should be a cap on its input and one on its output. The chips in that module do not work on 12v so there needs to be multiple small dc to dc power supply of some sort in there.

[redwire]

First off is to check the power supply voltages before and after it croaks.
I can see three voltage regulator IC's which seems a bit excessive.

I see U2 LP2954 as the 5V regulator for the load-cell excitation CN5_E+ and perhaps the analog circuits. Should be +5V across C13.
U1 looks like another voltage regulator, I would also check across C20. If there is a problem with the two tantalum capacitors, one can short and a regulator should get hot.
I'm not sure why a +6V regulator is needed, it might be a pre-regulator or it's to power a solenoid or something for powder dispensing? Where does that connect to? Connector CN4 goes to what?
I'm wondering if it has a sleep mode that is malfunctioning, or maybe adjustable LCD contrast that is a problem, or just the power supply is fizzling out.
What is SWA1 under the buzzer doing? A shake or touch switch or something?

[redwire]

I noticed some people found a problem on the dispenser (motor) board https://nictaylor00.blogspot.com/201...ad-defect.html
Maybe post pics of it.

[TortureGarden]

Start with that 7806 6V regulator. See. These there is sufficient power getting to it and from it. There also should be a cap on its input and one on its output. The chips in that module do not work on 12v so there needs to be multiple small dc to dc power supply of some sort in there.

Checking the PSU plug, I get 9.19V, which is the same as the voltage that I get on the 6V regulator input. At the output of the regulator, I get 6.04V.

[TortureGarden]

First off is to check the power supply voltages before and after it croaks.
I can see three voltage regulator IC's which seems a bit excessive.

Reading of 9.19V before and after

I see U2 LP2954 as the 5V regulator for the load-cell excitation CN5_E+ and perhaps the analog circuits. Should be +5V across C13.

Yep, dead-on 5.00V across C13

U1 looks like another voltage regulator, I would also check across C20. If there is a problem with the two tantalum capacitors, one can short and a regulator should get hot.

Across C20, I get 2.50V. None of the regulators are at all hot.

I'm not sure why a +6V regulator is needed, it might be a pre-regulator or it's to power a solenoid or something for powder dispensing? Where does that connect to? Connector CN4 goes to what?
I'm wondering if it has a sleep mode that is malfunctioning, or maybe adjustable LCD contrast that is a problem, or just the power supply is fizzling out.
What is SWA1 under the buzzer doing? A shake or touch switch or something?

The unit is two separate components; the powder dispenser and scale are stand alone units (as much as the dispenser requires the scale as a controller). As per the URL you added, the dispenser has its' own PCB that I fortunately, have not had an issue with.

The 6V regulator appears to connect to C14 then C12 then C31 then C9 then to U2, for what it's worth. CN4 connects to the pins that output to the connector for the powder dispenser. I have no idea about SW1, it's not something that is accessible when the unit is assembled. It's a manual 2-position switch which is labelled "ADJ/LOCK". It was in the LOCK position when the unit was disassembled.

[CapLeaker]

TortureGarden
Maybe try freezing some IC's with a can of air upside down? When the display fizzles out, the regulators stay on? Do the crystals keep working too? Maybe a problem then with the MCU?

[TortureGarden]

TortureGarden
Maybe try freezing some IC's with a can of air upside down? When the display fizzles out, the regulators stay on? Do the crystals keep working too? Maybe a problem then with the MCU?

Haven't had a chance to give the IC-freezing a go yet, will get some on the weekend and report back. The regulators do stay on when the LCD craps-out. I forgot to mention that when the unit is powered-on, it goes through what appears to be a self-test sequence and when the LCD was functional, the characters would display counting down from 0 to 1 and then audibly beep. It still does this sequence sans LCD output - assume this means that the MCU is still functioning in some capacity?

[redwire]

Oh it's a segment LCD display. It acts like the LCD backplane is not being pulsed or has an open connection. I wonder if Q12 does that? Q11 turns on the backlight LED.
It's a bit weird the MCU uses two crystals, no RTC that I can see so maybe it's for low speed LCD driver? I would try touching the leads on 32.768kHz crystal X2 to see if the oscillator is dying and needs a bump. Or try a gentle tap to the crystal.
Beyond this, taking the LCD apart to see if there are bad connections, solder balls in the zebra strips. That's a bit of surgery.

There are lots of complains about the Chargemaster 1500 not being reliable and they seem to just replace the board.

[TortureGarden]

So, not being near as savvy as you guys and having only rudimentary skills and equipment, I am beyond appreciative of your helping me out this issue. Thanks heaps.

With my el-cheapo multimeter on diode mode (I think this is what I read as a very basic means of testing a transistor) I tested the collector and emitter sides of Q12 (it has "CR" on it). The readings I got mean nothing to me but to those in the know, it may; one side, I assume the collector, I got a reading of 1577 and then lost the reading altogether. Tested several times and the same thing happened. Other side if Q12, I get a steady eading of 698. Thinking the disappearing reading on Q12 odd, I found that Q10 also appears to be "CR" transistor so, I tested it also. Got steady readings of 1568 and 690. Does this meant that potentially, Q12 is cactus?

Unfortunately, I have not got an oscilloscope to test the crystals but I did give them a gentle love-tap to no avail.

Prior to this, I disassembled the LCD, inspected it under the microscope (looks fine) and cleaned all of the contact points (LCD & zebras).

Really, I have no great idea of what I'm doing but willing to give it a crack.

My Chargemaster has served me flawlessly for many a moon and it would be sad to say goodbye. I think that RCBS no longer offer any support for this product and now point punters to https://lonemountaincommunications.com/. I might send them an email and ask if they have any schematics that they're willing to share.

[CapLeaker]

What is that header on the bottom left saying flash /user? Is this a TTL port? If everything is keeps working, there should be no problems if you hook a TTL to that port and look at it with a terminal screen. If you see a kernel panic, well that's a problem.

[TortureGarden]

I sent the "official" repair centre an email and this is the response they sent me.

[TortureGarden]

What is that header on the bottom left saying flash /user? Is this a TTL port? If everything is keeps working, there should be no problems if you hook a TTL to that port and look at it with a terminal screen. If you see a kernel panic, well that's a problem.

How do I determine if it is a TTL port?

[redwire]

I sent the "official" repair centre an email and this is the response they sent me.

I don't see a dedicated LCD controller IC. The main MCU H8 looks like it does it, look at the >26 PCB traces from it to going under the LCD display. I wonder if they mean that IC (MCU) is pooched and unobtainium. Hitachi always was quick to discontinue semi's.
Or that support email is full of malarkey.

[CapLeaker]

I thought about that. That's why I went with the TTL idea to see what's what.