Disappointed - Manual scheduled dosing limit
As a programmer and hardware developer with 20 years of experience, I feel compelled to share my recent experience with the GHL Doser, which left me both surprised and disappointed. This background becomes relevant as you read on.
The Problem
Last weekend, I discovered AEFW (Acropora Eating Flatworms) in my tank—a problem that was entirely my own fault for trusting a local hobbyist without proper quarantine. To address it, I decided to add two wrasses for biological control, even though they came from ich-infected tanks at my LFS. Having successfully treated ich in the past using the hydrogen peroxide dosing protocol outlined on the Humblefish forums, I wasn’t overly concerned about the ich. My focus was on eradicating the flatworms.
In the past, I used a Kamoer doser for peroxide dosing. However, I opted for the GHL Doser this time, believing its technical capabilities would better suit the complex dosing schedule required for ich treatment. Here's the specific peroxide dosing protocol for a 130 gallon tank, tailored to target the parasite's life cycle:
| # |
Time |
Volume |
|---|
| 1 |
9:00 |
26ml |
| 2 |
15:00 |
26ml |
| 3 |
21:00 |
26ml |
| 4 |
0:00 |
2.16ml |
| 5 |
0:15 |
2.16ml |
| 6 |
0:30 |
2.16ml |
| 7 |
0:45 |
2.16ml |
| 8 |
1:00 |
2.16ml |
| 9 |
1:15 |
2.16ml |
| 10 |
1:30 |
2.16ml |
| 11 |
1:45 |
2.16ml |
| 12 |
2:00 |
2.16ml |
| 13 |
2:15 |
2.16ml |
| 14 |
2:30 |
2.16ml |
| 15 |
2:45 |
2.16ml |
| 16 |
3:00 |
2.16ml |
| 17 |
3:15 |
2.16ml |
| 18 |
3:30 |
2.16ml |
| 19 |
3:45 |
2.16ml |
| 20 |
4:00 |
2.16ml |
| 21 |
4:15 |
2.16ml |
| 22 |
4:30 |
2.16ml |
| 23 |
4:45 |
2.16ml |
| 24 |
5:00 |
2.16ml |
| 25 |
5:15 |
2.16ml |
| 26 |
5:30 |
2.16ml |
| 27 |
5:45 |
2.16ml |
The GHL Doser Issue
When setting up the GHL Doser, I discovered its hard limit of 8 manual doses. While it supports over 96 automated scheduled doses, these cannot be edited after being generated, rendering the feature nearly useless for a precise, tailored dosing schedule.
I tried to work around this limitation by using two dosing heads and 2 tee fittings (one on input from peroxide to the 2 dose heads, and one output from the dose heads to a feed line to tank) to separate day and night dosing schedules. Unfortunately, the automated scheduler for the nightly dosing only allows a start time—there’s no option for an end time or more control over dosing intervals (other than daily variations).
Potential Solutions
Here’s how I’ve considered addressing this issue:
- Use all 4 dosing heads: Tee them together at the input and output, and assign the 8 manual scheduled times of each doser to cover the 27 required doses. Consuming all 4 heads of the GHL Doser just for this task seems wasteful and expensive.
- Build a custom controller: I could use an Arduino or ESP32 to simulate “open contact/closed contact” signals to the GHL dosers via the sensor port. However, this defeats the purpose of purchasing an advanced doser. If I’m going this route, I might as well use a $40 Kamoer stepper motor and dosing head and build my own doser.
Broader Frustration
What’s most disappointing is that this issue has been raised repeatedly since at least 2017. There are dozens of threads online and reports of direct feedback at trade shows, but GHL continues to dismiss this as a low-priority feature. As a developer, I can confidently say this would take no more than an hour of work—most of the code is already in place. Simply enabling an "edit" button for the automated schedule would resolve the problem. Moreover, what’s the point of having a start time in the automated scheduler if there’s no stop time?
Even budget Kamoer dosers from China offer these features and seem more responsive to customer feedback and requests.
Final Thoughts
I wanted the GHL Doser to be a versatile and precise solution, but it has fallen far short of my expectations. Has anyone else run into these limitations, and if so, how have you managed to work around them without spending more money for the profilux? GHL has some great hardware, but it’s clear they need to prioritize user needs more seriously.
