G-Box V3000 Fuse or Reset?

[Anole]

off the top of my head:

The diode is easy to replace, and Radio Shack probably does sell some that'll do.
(Ya might want to get a $5 soldering iron at Harbor Freight, instead of $8 at RS)
However, let's take a moment before using any 1N4000-series parts.
They are only rated at one amp.
If this is for the 36 volt side, I believe you should look for 3 amp or bigger diodes.
And it would be wise to ask why the diode blew, too.
If shorting your motor wires took it out, this would be the first I've heard of a chink in the otherwise excellent reputation of the Gbox.

serious edit after closer look:
A quick look at your picture suggests the bottom TWO rectifiers are blow'd up ! -
And just from the way the parts are laid out, I'd say the four smaller rectifiers under the white/yellow wires, are the 1N4000-something, and supply 12v power.
The bigger rectifiers to the left of where the white/yellow wires go through the board, are a full wave bridge (likely) for the 36 v motor power.
Two are certainly gone! - Good job exploding them! - :rant:
So, if both of the two diodes blew, then something major shorted that half of the power supply, and the rectifiers acted as the fuses! -

I'd see if I could locate the cause of the failure.
Shorted wiring between Gbox and motor?
Bad or shorted wire clips at the back of the Gbox?

 

[phlatwound]

Thank you all for your thoughts and observations, much appreciated.

Mr. Sadoun PMed me and is going to review this thread and let me know what he thinks.

If he has never seen this issue caused by anything other than operator error I am willing to accept that I unknowingly did something wrong when hooking or unhooking while I was bench testing, I trust his judgment and experience.

And if that's the case I will be looking for some soldering lessons.

 

[Sadoun]

Yep, Anole is right. The two rectifiers in your picture look like they are shot.

Here is a picture of the diode bridge rectifier: It is used to convert AC to DC.



This is my guess on what could have caused the damage to these rectifiers:

When using the GBOX to provide the current needed to start a linear actuator, there is always the potential for some sparking (if the GBox is ON when you connect the motor wires on the back of the GBox). Sparking indicates that the flow of current has been interrupted and then restarted. In any circuit in which the load has inductance, there is a great potential for damage to occur from sparking.

An inductive load, such as an electric motor, generates a magnetic field. When the flow of current is interrupted suddenly, the magnetic field collapses. The energy in the magnetic field induces a voltage in the windings of the motor. It is very possible that when this occurs a voltage spike will be generated. The voltage in the spike can be greater than the exciting voltage (the 36-volt) which operates the circuit.

These harmful voltage transients could be generated. If the level of these transients exceeds the maximum reverse voltage junction rating of any rectifiers in the circuit, those rectifiers will be damaged.

My recommendation is to never connect a GBox to a motor before turning it OFF 1st and disconnecting it from power. Same as with with multi-swicthes, etc.

I will consult with factory engineer on this as well to check his opinion.

 

[Sadoun]

Radar made a good suggestion on how to change the diodes (rectifiers). That is how this circuit will have to be fixed. If you don't feel comfortable doing that, send the board to us and we will fix it for you.

 

[pendragon]

However, let's take a moment before using any 1N4000-series parts. They are only rated at one amp. If this is for the 36 volt side, I believe you should look for 3 amp or bigger diodes.

Absolutely correct! At the barest minimum you need something like a 1N5402 or 1N5404. I vaguely recall the power supply being rated around 3A, so a 3A diode is arguably a mediocre fuse. Another possibility is to wire in a 4-5A full-wave rectifier instead of the diodes.

And it would be wise to ask why the diode blew, too.
If shorting your motor wires took it out, this would be the first I've heard of a chink in the otherwise excellent reputation of the Gbox.

The GBOX has a better pulse resolution than other models and may have a more reliable pulse detection circuit. However it is VERY cheaply built inside. The motor/sensor connector is a complete joke. I am not impressed by it, but there's not a lot out there to choose from.

And just from the way the parts are laid out, I'd say the four smaller rectifiers under the white/yellow wires, are the 1N4000-something, and supply 12v power. The bigger rectifiers to the left of where the white/yellow wires go through the board, are a full wave bridge (likely) for the 36 v motor power.

You are correct about the layout. There is absolutely NO +36V protection circuitry at all in the GBOX other than the ratings of the diodes. They are probably counting on the power supply surviving a short long enough for the pulse detection circuit to realize that nothing is coming back. Not the greatest idea.

Two are certainly gone! - Good job exploding them! - :rant:
So, if both of the two diodes blew, then something major shorted that half of the power supply, and the rectifiers acted as the fuses! -

Good job spotting the problem. I had eye surgery on Friday and I'm not even supposed to be on my computer. So I didn't look at the pictures. I'm kind of working from memory and some tear-down notes I took when my first two GBOXes arrived in unusable condition (both transformers had ripped away from their mounts in shipment and had trashed all the circuit boards. I repaired enough of the traces to get them working, but it may now be obvious why I am only a lukewarm supporter of these units).

I'd see if I could locate the cause of the failure.
Shorted wiring between Gbox and motor? Bad or shorted wire clips at the back of the Gbox?

Absolutely. I would check the motor/sensor connector because it is so flimsy. I suppose the east/west relays could have turned on at the same time, which would cause nasty consequences. It's also possible that one or more relay contacts are welded. You need to check this through carefully before applying power. Or send it back.

 

[pendragon]

Yep, Anole is right. The two rectifiers in your picture look like they are shot.

Here is a picture of the diode bridge rectifier: It is used to convert AC to DC.

There is no regulator in the +36V circuit if my notes are correct. I would have really liked to see a current-limited regulator in this type of circuit because the potential for shorts is so high.

An inductive load, such as an electric motor, generates a magnetic field. When the flow of current is interrupted suddenly, the magnetic field collapses. The energy in the magnetic field induces a voltage in the windings of the motor. It is very possible that when this occurs a voltage spike will be generated. The voltage in the spike can be greater than the exciting voltage (the 36-volt) which operates the circuit.

These harmful voltage transients could be generated. If the level of these transients exceeds the maximum reverse voltage junction rating of any rectifiers in the circuit, those rectifiers will be damaged.

My recommendation is to never connect a GBox to a motor before turning it OFF 1st and disconnecting it from power. Same as with with multi-swicthes, etc.

I will consult with factory engineer on this as well to check his opinion.

This is possible, but I would hope the diodes are rated somewhere from 200-400 PIV which should minimize this kind of problem. A dead short is an easier explanation given the lack of any real protection circuitry. Either way, if you were bench testing this with short cables, the potential for disaster would be much higher than with a long cable.

 

[Sadoun]

A dead short is an easier explanation given the lack of any real protection circuitry.

Yes, dead shorting the terminals could do that as well. OP, please check your motor wires for short, corrosion, or for any short somewhere in the motor.

 

[Anole]

some improvements:

In the hardware I've designed for production, I always try to make it idiot proof.
And as the old joke goes: the problem is... they're always coming up with better idiots! -

So, I'd have put in a 3 amp transformer, 5 amp diodes, and maybe a circuit breaker.
Hopefully, it should sustain a short long enough to trip the breaker.
Even one of those self-resetting breakers, would be okay... it's just money. -

Then, some arc/pulse/noise suppression/bypass on the motor connectors...
And maybe reverse protection diodes across the relays, or some other surge suppressors, as appropriate.
The box hits end-of-life the minute one of the relay contacts welds shut, so preventing that is big-time serious!
Maybe a low ohm surge resistor in series with the motor?

The spring clips to connect up the motor wires are a sore point. - :rant:
So maybe one of us should outfit ours with big five-way binding posts, shoot some pictures, and start a movement! -

 

[Sadoun]

Ok, Anole you have got some good points there. Let's start a new thread on ways we can improve the GBox V3000 design. I like to make the V3001 version more robust. A few dollars more, will make for a more serious product.

Let's roll.

 

[phlatwound]

Radar made a good suggestion on how to change the diodes (rectifiers). That is how this circuit will have to be fixed. If you don't feel comfortable doing that, send the board to us and we will fix it for you.

If I decide not to try it myself I will send it to you and you can send it back with my next order, thanks.

Yes, dead shorting the terminals could do that as well. OP, please check your motor wires for short, corrosion, or for any short somewhere in the motor.

This occurred while I was bench testing a motor with leads that were about 2' long, there is no issue with the leads.

All I can figure is the power leads shorted out when I was physically moving the G-Box (or motor) on the bench, but both leads only had about 1/4" stripped on the ends and that was inside the spring clips, the connections to the motor wires were made with wire nuts, I suppose it's possible that something got crossed there.

Motor is installed and running on my BV mount now, no problems there.

The only other possibility I can think of is what you said about connecting/disconnecting with the unit powered up.

I'm still confused as to what the fuse is that is mentioned on the website? After what I've learned here I'm guessing it is for the sensor circuit only? Can you point it out in my picture?

In the hardware I've designed for production, I always try to make it idiot proof.
And as the old joke goes: the problem is... they're always coming up with better idiots! - .......

I find this quite offensive and I thought I could expect better out of you, Anole!

The very implication that there MIGHT be a better idiot out there than me is UNCALLED FOR AND JUST PLAIN WRONG...I AM THE BEST IDIOT AROUND!!!!!!!!!!!!!!!!!!

 

[AcWxRadar]

The diode is easy to replace, and Radio Shack probably does sell some that'll do.
(Ya might want to get a $5 soldering iron at Harbor Freight, instead of $8 at RS)
However, let's take a moment before using any 1N4000-series parts.
They are only rated at one amp.
If this is for the 36 volt side, I believe you should look for 3 amp or bigger diodes.
And it would be wise to ask why the diode blew, too.
If shorting your motor wires took it out, this would be the first I've heard of a chink in the otherwise excellent reputation of the Gbox.

serious edit after closer look:
A quick look at your picture suggests the bottom TWO rectifiers are blow'd up ! -
And just from the way the parts are laid out, I'd say the four smaller rectifiers under the white/yellow wires, are the 1N4000-something, and supply 12v power.
The bigger rectifiers to the left of where the white/yellow wires go through the board, are a full wave bridge (likely) for the 36 v motor power.
Two are certainly gone! - Good job exploding them! - :rant:
So, if both of the two diodes blew, then something major shorted that half of the power supply, and the rectifiers acted as the fuses! -

I'd see if I could locate the cause of the failure.
Shorted wiring between Gbox and motor?
Bad or shorted wire clips at the back of the Gbox?

Thank you Anole,

This is a good point, I couldn't detect if the blown diodes were for the logic voltage supply or for the actuator power supply. The 1N400X series diodes are pretty tough general purpose diodes, but not rated for the 3A supply.

No problem though. Rat Shack or any fairly good electronics shop / outlet should have generic diodes on hand that will handle the power load needed. Be sure to look up the spec sheets before you buy them. You can usually get by just fine with most selections. There are many general purpose diodes that will fit the bill, so don't be too persnickety, just get the wattage or current capacity correct.

One other note, make sure you replace all four diodes in the bridge circuit at the same time. You won't know if the other diodes have been harmed by whatever damaged the others and cannot judge their life span. Also, it is a good policy to replace such components with all the same style.

You should also beware to check any electrolytic filtering capacitors on the output side of the bridge rectifier. Electrolytic capacitors are often prone to shorting and that could very well have been what caused the diodes in the rectifier to blow out. If you witness any capacitor that is bulgy (bloated), discolored or shows signs of corrosion near its base (like battery terminal corrosion) I would highly suspect that capacitor to be bad. You can get replacement caps from Rat Shack, too.

RADAR

 

[Sadoun]

Radar,

You sound like you have an electronics engineering background? I am proud of you. This is what makes this place and nice hangout. My background is mechanical/manufacturing engineering (BSME) with a passion for electrical/electronics circuits (no expert there).

 

[AcWxRadar]

Radar,

You sound like you have an electronics engineering background? I am proud of you. This is what makes this place and nice hangout. My background is mechanical/manufacturing engineering (BSME) with a passion for electrical/electronics circuits (no expert there).

Hi Sadoun,

You are correct. I have a Associates Degree in Electronics Engineering. I worked for a company in Lincoln, Nebraska called Duncan Aviation for ten years. While there I worked on weather RADAR systems for private and corporate aircraft, hence my screen name here (AcWxRADAR).

I very much enjoy fooling around with anything electronics, but I have a host of other hobbies that cover all sorts of avenues.

One of the greatest learning tools that I was taught while learning my trade was this... You do not necessarily need to know everything off the top of your head or from memorization. There is too much information out there to store it all within the confines of our brains, so you just need to know how to research well! That is the key and the best tool anyone can have.

The second best tool to applying your knowledge is understanding how to divide and conquer. Not on a battlefiled, but within an electronic circuit. Check the power supplies first, as a common rule. Then divide and subdivide the circuitry into smaller blocks and continue to break them down until you isolate the defect.

I have spent 25 years or better developing a good sense for troubleshooting electronics. I feel that I understand it fairly well, but there are always these little quirks that boggle me. I find that troubleshooting computer programs present me with a nemesis. People have different styles of composing programs and sometimes these do not seem logical to me. You can easily get lost in someone else's programming.

If I had spent more time with this avenue, I might be able to identify other's mistakes, but as it is, I am simply just a greenhorn to programming and IT concerns. Now that I have this Linux OS receiver (the AZBox and the AB IP Box) I have to learn something new. I find it a challenge as I am a bit behind in this endeavor.

All of this is simply fun and food for life. Obviously it is not impossible, but it may be quite ellusive for a time. Here is where the challenge and the fun comes into play. And this is why I follow this hobby with great fervor! This is how I feel that I can learn something new. I feel that a person can never stop learning, or at least they cannot stop trying to.

This is why I enjoy this site so much! I learn a great deal from the members here. Hopefully I am providing some information in return.

Either way, it is a lot of fun!

AcWxRADAR (Gordy)

 

[Anole]

- good humor!

Originally Posted by Anole
In the hardware I've designed for production, I always try to make it idiot proof.
And as the old joke goes: the problem is... they're always coming up with better idiots! - .......

I find this quite offensive and I thought I could expect better out of you, Anole!

Hey, I just calls 'em as I sees 'em! -

The very implication that there MIGHT be a better idiot out there than me is
UNCALLED FOR AND JUST PLAIN WRONG...I AM THE BEST IDIOT AROUND!!!!!!!!!!!!!!!!!!

Well, I don't know 'em ALL ! - -

 

[mikelib]

You might consider installing one of these units.
Bridge Rectifiers | AllElectronics.com

 

[Anole]

Some of those bridges are certainly gonna survive a shorted output, when fed by a 3 amp rated transformer, and never even shrug. -
That's the kind of over-engineering I like to sneak into a product.
Just so long as manufacturing or the purchasing agent doesn't catch or question the decision.

As you see, the prices are quite reasonable.
And a $1, $2, or $3 part that just never gives up the ghost, sounds reasonable to us consumers.
Problem comes when the bean-counters decide to use a 10¢ part instead of the 25¢ unit you spec'd.
Or in this case, maybe 3¢ vs 5¢ -

 

[cruzin]

Almost same problem

I have been following this post as I have a new gbox and a new actuator and have the same E2 error when moving one way. I do not find any blow parts in mine though.
What I do find is on my 2 white wires I have 14vdc and at the relays I have 49.9-50 vdc across 2 of the pins on both relays.
When I press the left button to move (farthest from on/off) I get both relays feel like the click. But when I press the button next to on/off I get no click.
I was wondering why I have 49 to 50 vdc at the relays.

I checked voltage at diodes and I have voltage there like telling to check.
Any suggestions?

 

[pendragon]

When I press the left button to move (farthest from on/off) I get both relays feel like the click. But when I press the button next to on/off I get no click.
I was wondering why I have 49 to 50 vdc at the relays.

I checked voltage at diodes and I have voltage there like telling to check.
Any suggestions?

If you have 49-50V at the relays and you hear no click going one direction (west?), you have a different problem than OP. Possibilities are all related to why the relay doesn't engage, such as a bad connector, broken or shorted trace, bad relay driver, jammed relay or burned-out relay coil. You may want to check to see if you're getting any voltage on the relay coils when you drive east vs. west (on the bad side you'll only get a short blip).

 

[cruzin]

e-2 and voltage

`Hi, I checked it. And I get 49vdc when pressing the button closest to the display and no voltage when I press the button next to the on/off button. I would like to know which of the terminals I could check for voltage that would cause the relay to close so if I am getting no signal to the relay to close it would be the cause I suppose. Why am I getting so high of a voltage 49 vdc.

 

[pendragon]

Get the vendor and part number off the relay and look up the pin-outs. That should tell you where to check.

I wouldn't worry much about the 50 V. I have the same. There is only a tiny capacitor on the full wave rectifier output and there is no way this can store enough energy to ride through the 120 Hz peaks when running a motor/actuator. If you were to put a scope on the output when driving a dish, the average voltage would be a lot less.