Late haze and sugar loss in filtered backsweetened cider

[kalp patel]

Hello everyone,

I'm looking for help diagnosing a stability issue in our commercially produced backsweetened cider.

We have been using the same process since September 2023:

• ~6% ABV

• pH ~3.7

• Residual sugar 20–25 g/L

• 0.45 µm absolute filtration 3m

• Potassium sorbate

• KMS targeting ~25 ppm free SO₂ (70–90 ppm total SO₂)

• No post-packaging pasteurization

All batches produced from September 2023 through March 2025 remain clear and stable. However, more than 50% of batches produced after March 2025 have developed the following symptoms after 3–4 months of storage at ambient temperatures in India (35–38°C):

• Turbidity/haze and visible sediment

• Noticeably more acidic/sour taste

• More prominent malic character

• Little or no aroma change

• No excessive carbonation, gushing, or bottle bombs

In one affected batch, residual sugar dropped from ~24 g/L at packaging to ~18 g/L after storage. pH in affected bottles measures around 3.6 versus 3.7 at packaging, although this may be within measurement error.

Additional information:

• The concentrate used for backsweetening was plated and showed no growth.

• Packaged cider passes our heat and cold stability tests before release.

• Retained bottles from November 2023 are still clear and taste as expected.

We have a small in-house microbiology lab and routinely plate packaged cider. We typically see:

• 5–10 CFU/100 mL yeast colonies

• Occasional white raised colonies on WL Nutrient Agar

Even after laboratory pasteurization of some samples, we occasionally recover 1–3 CFU/100 mL. We currently have limited ability to identify the organisms.

As a precaution, we have strengthened sanitation (0.5% caustic, citric rinse, 450 ppm PAA, and PAA bottle rinse), but I am still concerned about recurrence.

My questions are:

1. Does this sound more like low-level yeast refermentation, bacterial spoilage, or something else?

2. At pH 3.7 and 25 ppm free SO₂, would you consider this preservation regime adequate for cider stored at 35–38°C?

3. Are 5–10 CFU/100 mL in packaged cider a concern, or are low counts like these sometimes considered acceptable?

4. Should packaged cider ideally plate completely clean , especially after 0.45 µm absolute filtration?

5. Given that the process worked for ~18 months before these issues appeared, where would you focus the investigation first?

Any thoughts would be greatly appreciated.