Apex conversion to GHL

[Lasse]

Has anyone figured out how to drive a small 12v load with GHL. I see that the breakout box shows two relay connections for the L Port but I can't find any documentation on how to control them.

Whats you mean with small 12 V load - how man mA?

Sincerely Lasse

 

[scolson]

Whats you mean with small 12 V load - how man mA?

Sincerely Lasse

Something like a 12V solenoid. I haven't checked the current yet on that yet. If it is too much I could stick an opti isolated relay in to minimize the load back to the controller. It looks like it is possible, but that the relay ports on the L ports are tied to the 1-10V signal state so I'd loose the 1-10v functionality. I was hoping they were independent.

I have a 4e, so no S ports unless I buy an expansion card.

 

[theatrus]

Awesome thread! Thanks for taking the time to do the comparison, Bill. I got an Apex about 3 years ago and have been happy so far. I was going back and forth between it and GHL. At the time I had a very poor response from GHL when I had questions so that, combined with the fact that I don’t own a Windows computer and that GHL was going to be at least twice as much for my needs pushed me to Apex.

After reading through all 13 pages I had a few thoughts:

I was frustrated by my salinity probe when I got my apex. I played around with it, etc and eventually forgot about it. Several months later, I looked at it and found it was reading very stable and consistently. What I really realized, though, was that the salinity probe is not necessary for my system. I have a 120 w/a 40 breeder sump, estimated total volume ~130gallons. My ATO tank is 10 gallons. Even If the entire 10 gallons of DI water got dumped into my DT the SG would drop from 1.026 to 1.024. Not ideal but not enough to cause major damage. This doesn’t excuse the finickyness of the Apex probe, but I think people worry too much about it.

Regarding the power bar/relay capacity - Does anyone know what the capacity of the Apex relays actually is, or the GHL relays? I’ve looked online for a teardown of either device and can’t find anything. The reason I ask is the current capacity is determined not just by the relays but also by things like the tracings on and other circuitry on the PC board, cooling, etc. It’s possible that Apex actually uses 15A relays but rates them at 7A because of the board design, or that they have already downrated them in their specs and 7A is really 75 or 80% of the actual capacity.

I have no actual data on either company’s power bars, so this is all speculation, but I did want to point out that you were actually making some assumptions about both devices without actually knowing.

I’m interested in a comparison of the flow meters (as well as how GHL’s meters actually work. If they do work by measuring vibrations due to turbulence they may also be susceptible to buildup/accumulation on the sensor, since that could affect the frequency of vibration. Again, this is hypothetical; I haven’t seen any actual comparison data between either.

Finally, regarding BRS, I‘m a supporter, but I also share some of the concerns regarding their recent mergers and some of the same observations with their reviews. I distinctly remember a few years ago when the Vertex Cerebra was announced - they were big proponents in their videos but they really failed to do any sort of critical evaluation and basically parroted Vertex’s talking points.

They’ve carried Apex gear for years but only recently started carrying GHL. Was this because of GHL’s relatively small market share? Incentive payments by Neptune? simply a business decision on their part? Since they were practically promoting the Cerebra initially I don’t think their relationship with Neptune was terribly close at that point but who knows? Regardless, I think future controller reviews by BRS will need to be taken with a bit of skepticism since they will have a conflict of interest.

Relay shots:

 

[Lasse]

You need a relay - but it is true - on the L ports the 12 V signal is depended of the 1-10 V signal. You can trick it with a false (virtual) ill curve but you can´t use that L channel as an independent 1-10 V source

Sincerely Lasse

 

[Sleepydoc]

Thanks, for the pics, @theatrus, that’s what I was looking for. The relays in the EB832 are actually 250V/16A relays, so they are actually running well below their specs. I’m guessing it’s the tracings on the circuit board that limit the current to 7A. Either way, I don’t don’t want to further derail @Billldg ’s excellent thread.

 

[theatrus]

Thanks, for the pics, @theatrus, that’s what I was looking for. The relays in the EB832 are actually 250V/16A relays, so they are actually running well below their specs. I’m guessing it’s the tracings on the circuit board that limit the current to 7A. Either way, I don’t don’t want to further derail @Billldg ’s excellent thread.

From inspection, limits are: current measurement shunts, PCB tracks, general derating. It appears to be a 1oz board (though did not measure, just a rough comparison to some 2oz boards), and track widths were not maximized on the bus. Its all a bit intermingled with the PSOCs doing the power measurements at each outlet, since its all carried on the same PCB for cost reasons (though at the same time there are two power supplies sandwhiched into the strip for what I imagine are purely engineering efficiency and change reduction reasons, and not cost, which is a whole other topic on the electic build refinement or lackthereof on Neptune products).

Running more than 10A on any outlet strip is a bad idea anyway unless its a 5-20 plugged variant, so its all somewhat a moot point.

Excited to see the GHL alternatives, especially for the Trident replacement (which is the only thing I use my APEX for).

 

[Billldg]

Doing my homework before I install the flow sensor, I just read that you can't add 2 flow sensors to a PL-LY splitter cable. it has to be directly connected to the P4 or expansion card.

 

[Billldg]

I just hooked up the flow sensor, and WOW, I am surprised of how much loss I have on a Sicce 4.0. It is rated for 950 gph, I knew I would have some loss, but, I guess the UV sterilizer takes even more than I thought. I am flowing roughly 350-360 gph. No where near enough for what I need the UV sterilizer to do for Dino's. Hooking it up and programming it was very simple. Below is my new Dashboard:

Below is when I programmed it and the link to the tutorial on programming it.

How to setup GHL Flow Sensors

www.aquariumcomputer.com

 

[Billldg]

Actually, I missed the Calibration factor step...OPPPSSS!!! ;Facepalm;Hilarious

Now the flow is around 690-720. That is kinda what I was expecting...AND hoping for, LOL!!!

Gotta love Sicce pumps!!! I knew it wouldn't be that far off.

 

[Sleepydoc]

Have you calibrated your sensor and/or done a bucket flow test? I wouldn't expect it to be off that much but it would be interesting to see. The Sicce 4.0 is only rated to 12 ft and according to the spec sheet it wold flow about 350 gpm at 9ft, so your numbers are at least believable.

Edit - I just saw your f/u post. Glad it's working better. Calibration is important!

 

[Billldg]

Have you calibrated your sensor and/or done a bucket flow test? I wouldn't expect it to be off that much but it would be interesting to see. The Sicce 4.0 is only rated to 12 ft and according to the spec sheet it wold flow about 350 gpm at 9ft, so your numbers are at least believable.

^^, I missed the calibration step.

 

[Billldg]

Also, you can only add 4 flow sensors to a P4. This is a little concerning to me personally, but not most, as I will use 4 flow sensors on the tank itself. Adding the UV sterilizer flow sensor makes 5, thus the concern. 80 percent of reefers will not use any more than 2 flow sensors, I am the exception...LOL!!!

 

[KStatefan]

Actually, I missed the Calibration factor step...OPPPSSS!!! ;Facepalm;Hilarious

Now the flow is around 690-720. That is kinda what I was expecting...AND hoping for, LOL!!!

Gotta love Sicce pumps!!! I knew it wouldn't be that far off.

I am surprised you are getting that much. I would have thought the 1/2" out of the sensors would have restricted it more.

 

[Billldg]

I am surprised you are getting that much. I would have thought the 1/2" out of the sensors would have restricted it more.

I think the 2377 flow sensor is a bit wider, but, don't quote me on that.

GHL list it as a 1" ID, but then, why restrict it to a 3/4" union?

It has me actually looking at buying a 3D printer to fabricate unions.

It is roughly 10 ft of 1" tubing, the UV sterilizer, and the flow sensor.

 

[KStatefan]

I think the 2377 flow sensor is a bit wider, but, don't quote me on that.

GHL list it as a 1" ID, but then, why restrict it to a 3/4" union?

It has me actually looking at buying a 3D printer to fabricate unions.

It is roughly 10 ft of 1" tubing, the UV sterilizer, and the flow sensor.

That is why I made my fittings.

I did not pay attention to your pictures and just saw the 6.22 ml/pulse figure that must have been prior to your change.

 

[Billldg]

That is why I made my fittings.

I did not pay attention to your pictures and just saw the 6.22 ml/pulse figure that must have been prior to your change.

I knew someone made fittings and referred me to a R2R member that made 3D unions, just wasn't sure as to who yet, as I have not looked back to see. I may buy a 3D printer to make the unions, UGH, but if it will help others, so be it. That's what this thread is for, to help other reefers. I already spent $3000+ to do the comparison, and the cost is growing, so reefers can get an honest opinion, what's a few more thousand...LOL!!! ;Hilarious ;Hilarious ;Hilarious

The whole point is we need to let GHL know what can make their controller better, so they can sell more controllers and get a bigger share in the US market. The bigger they become in the US, the more availability we will have.

 

[ingchr1]

Relay shots:

While not the 6E, here's the inside of a Powerbar 5.1. The date on it is 12/2016.

 

[Sleepydoc]

Nice pics - unlike the Neptune EB832, the outlets on the GHL power bar has the outlets mounted to frame and connected with wires via spade connectors vs soldered directly on the board. That puts less stress on the board when things are plugged/unplugged and cracked solder joints can be a source of failure. On the flip side the space connectors can potentially corrode in a humid, saltwater environment (that would be an easier fix than a broken solder joint, though.) I haven't heard of either of these being significant problems, so I don't think that's an issue.

The Relays are 250V 16A relays, just like in the EB832. Neptune uses Golden relays, GHL's are made by Hongfa - I have no idea which one is better. Either way, they're adequately sized and actual relay failures haven't been an issue with either device.

The second to last photo appears to be a 12 V power supply, likely for driving the relays. The 12V PSU has been a source of failure in the EB832s and Neptune has changed it in the newer versions. GHL's is a different, larger board. I see some Meanwell components which are generally high quality. The capacitors look very similar to the cheap capacitors that tended to fail in the older Neptune boards, but I can't see the brand. It wouldn't make sense to use good Meanwell components and then cheap out on the capacitors, though. The larger board tells me they probably used larger components giving more overhead.

One of the complaints I've seen of the Neptune gear (justified, IMO) is that it isn't conformally coated to protect against corrosion. It's hard to tell, but it doesn't look like the GHL board is either, unfortunately.

Overall, the GHL power bar seems to be well constructed from what I can see. A lot of the transistors have metal caps for heat dissipation and there is plenty of space for airflow around them.

One thing that's really hard to get over is the orange plastic of the EB832 - it just looks cheap. I haven't seen the GHL in person. From the pics it looks like it's plastic, too, but it doesn't look nearly as cheap as the Neptune.

 

[theatrus]

The Relays are 250V 16A relays, just like in the EB832. Neptune uses Golden relays, GHL's are made by Hongfa - I have no idea which one is better. Either way, they're adequately sized and actual relay failures haven't been an issue with either device.

No strong data on each. Its not a well known brand, but most manufacturers won't splurge on a TE, Omron, etc. It has all the safety marks.

The capacitors look very similar to the cheap capacitors that tended to fail in the older Neptune boards, but I can't see the brand.

Lots of SMD can types look very similar. I can't tell manufacturer from these (the top datecode marking might tell more, or indicate the series for Nichicon or Panasonic types). Most seem to be in a decoupling application, though a few are around the DC/DC stage in the last image.

One of the complaints I've seen of the Neptune gear (justified, IMO) is that it isn't conformally coated to protect against corrosion. It's hard to tell, but it doesn't look like the GHL board is either, unfortunately.

This is the biggest one for me. I've had to repair Neptune gear due to salt intrusion (amongst other things). Its a big process cost add step, but over time salt gets everywhere, and power strips will never be sealed unless you're routing cords through gaskets. Kessil for example fully coats and partially pots the control boards in the lights I've torn down.

A lot of the transistors have metal caps for heat dissipation and there is plenty of space for airflow around them.

Not seeing any of the transistor drivers, but this really isn't a problem spot. A simple multi-channel PFET or NFET chip will handle switching just fine, doesn't get hot, and will last forever. Prior Neptune strips were mostly based on a TRIAC design, and while silent and didn't suffer from mechanical wear, probably didn't have enough snubbing and eventually died an induced voltage death.

One thing that's really hard to get over is the orange plastic of the EB832 - it just looks cheap.

I have many gripes with modern Neptune, but their love affair with the latest APEX designs in doing massive plastic injections is just an epic waste of resources and money and I wish they'd knock it off.

I really don't need my power strip to have multiple colors of surround plastic when its not needed. The main controller is a giant blob of plastic for what amounts to an awkward 2-3 board stack of small PCBs. The Trident's top fin makes it hard to fit on a small shelf, and all it has in it is a light. Nothing is sized the same so it doesn't stack together. I preferred Neptune when they were just using COTS plastic enclosures machined for some holes and a sticker on top (like some of their modules are still) - it was amazingly boring, but functional. Someone injected some capital and they went custom plastics crazy (the plastics are cheap, the tools for a low volume product are a big NRE).

 

[vangvace]

The spades also let GHL use the same frame for all the power regions they service. I assume the inner brains are mutivoltage.

If I remember I'll open up a newer 6e. It wouldn't be till the end of the month. Maybe I will ask my lfs if they have a blown up 832 I can disassemble too