<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Tongda Machinery</title>
    <description>The latest articles on DEV Community by Tongda Machinery (@tongdamachinery).</description>
    <link>https://dev.to/tongdamachinery</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4211336%2F39e962a5-163a-464e-acec-0c340b95ca52.png</url>
      <title>DEV Community: Tongda Machinery</title>
      <link>https://dev.to/tongdamachinery</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/tongdamachinery"/>
    <language>en</language>
    <item>
      <title>Sizing a Blow Molding Line From Real Output, Not the Model Name</title>
      <dc:creator>Tongda Machinery</dc:creator>
      <pubDate>Mon, 09 Nov 2026 04:08:21 +0000</pubDate>
      <link>https://dev.to/tongdamachinery/sizing-a-blow-molding-line-from-real-output-not-the-model-name-2ke7</link>
      <guid>https://dev.to/tongdamachinery/sizing-a-blow-molding-line-from-real-output-not-the-model-name-2ke7</guid>
      <description>&lt;p&gt;When engineers size an extrusion blow molding line, the first trap is treating the litre figure in a model name as a production rate. It is not. On a Tongda machine—and on most industrial EBM builders—the model name follows the &lt;strong&gt;maximum container volume&lt;/strong&gt; that machine is built to make. It does not promise containers per hour.&lt;/p&gt;

&lt;p&gt;Tongda has supplied extrusion blow molding machines since 1999, with more than 10,000 machines delivered to over 100 countries and a range from 10 mL bottles to 3,000 L water storage tanks. The notes below are for process and project engineers who need a defensible machine count and a utilities plan before the purchase order is cut.&lt;/p&gt;

&lt;h2&gt;
  
  
  Three figures that actually describe output
&lt;/h2&gt;

&lt;p&gt;Ask the builder to quote these three items against &lt;strong&gt;your&lt;/strong&gt; drawing and &lt;strong&gt;your&lt;/strong&gt; mold—not against a brochure example:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Extruder output&lt;/strong&gt;, in kilograms per hour
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Dry-cycle time&lt;/strong&gt; for that machine configuration
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Accepted pieces per hour&lt;/strong&gt; on your mold
&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;A dry cycle is not accepted production. Kilograms per hour are not pieces per hour. Pieces per hour only mean something when they are tied to the mold you will run, because cooling, flash, resin, and the operator’s pace all sit between a volume class and the containers you can pack.&lt;/p&gt;

&lt;p&gt;Tongda does not publish a capacity table that turns a model name into pieces per hour. Published model pages, where they exist, are the place to read a figure that belongs to that model. Everyday output is not a fixed share of anything in the name. The full framing is here: &lt;a href="https://www.tongdamachine.net/machine-sizing-real-output/" rel="noopener noreferrer"&gt;Real output versus the model name&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  An illustrative sketch (example numbers only)
&lt;/h2&gt;

&lt;p&gt;The arithmetic below uses &lt;strong&gt;clearly labeled example numbers&lt;/strong&gt;. They are &lt;strong&gt;not&lt;/strong&gt; Tongda ratings, &lt;strong&gt;not&lt;/strong&gt; a quotation, and &lt;strong&gt;not&lt;/strong&gt; tied to any Tongda model. Replace every blank with figures quoted for your mold.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;EXAMPLE ONLY — not a Tongda specification

Plant need:              1,200 accepted pieces / hour
Quoted on your mold:       400 accepted pieces / hour
Machines to discuss:     1,200 ÷ 400 = 3 machines

Check against resin:
  Piece weight (example):     180 g = 0.180 kg
  Line throughput (example):  1,200 × 0.180 = 216 kg/h
  Compare to quoted extruder kg/h for the configuration you chose.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;If the plant need is higher than one machine can deliver on that mold, you discuss a second machine, a multi-head or double-station configuration, or a different cavity layout—not a larger litre number in the model name. The litre class only tells you the machine can physically form that volume; it does not size the line.&lt;/p&gt;

&lt;p&gt;A useful internal checklist before you send the enquiry:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Accepted pieces per hour on your mold: ______
&lt;/li&gt;
&lt;li&gt;Accepted pieces per hour the plant needs: ______
&lt;/li&gt;
&lt;li&gt;Machines to discuss: needed ÷ accepted on that mold
&lt;/li&gt;
&lt;li&gt;Extruder kg/h quoted for the configuration: ______
&lt;/li&gt;
&lt;li&gt;Dry-cycle time quoted for the configuration: ______
&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Leave the blanks empty until the builder fills them for your product.&lt;/p&gt;

&lt;h2&gt;
  
  
  Plant utilities: what to confirm before the machine arrives
&lt;/h2&gt;

&lt;p&gt;A blow molding line needs power, compressed air, and cooling water or a chiller before it can run in your plant. Ancillary choices—dryer, crusher, loader, compressor—are sized with the selected model. They are not one size for every machine.&lt;/p&gt;

&lt;h3&gt;
  
  
  Voltage and frequency
&lt;/h3&gt;

&lt;p&gt;Tongda &lt;strong&gt;customizes voltage and frequency to the buyer’s plant power supply&lt;/strong&gt;. Build the machine for the supply you have; do not assume a retrofit later. Send the plant voltage and frequency with the container drawing.&lt;/p&gt;

&lt;h3&gt;
  
  
  Air, chiller, and drying
&lt;/h3&gt;

&lt;p&gt;Compressed air (for blowing and for pneumatic movements), chiller capacity, and drying each &lt;strong&gt;depend on the machine model&lt;/strong&gt;. Tongda does not publish one air volume, one chiller size, or one drying-humidity threshold for the whole range. Say whether the resin is one you expect to dry, and whether a crusher should be in the quotation. Those answers are part of model selection, not a number on a general page.&lt;/p&gt;

&lt;h3&gt;
  
  
  Floor and material flow
&lt;/h3&gt;

&lt;p&gt;Before installation, confirm floor space, how the machine will be lifted in, how resin will be loaded, whether flash will be ground and returned, and how containers will leave the machine. Factory acceptance at Tongda’s plant is not the same event as site acceptance on your floor—utilities and downstream equipment at the plant can differ from the factory trial. Planning questions (without published kW or air tables) are collected here: &lt;a href="https://www.tongdamachine.net/site-utilities-planning-blow-molding/" rel="noopener noreferrer"&gt;Plant utilities for a blow molding line&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to put in the enquiry email
&lt;/h2&gt;

&lt;p&gt;A complete enquiry saves a round of clarification. Include:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Container drawing, resin, and target accepted pieces per hour
&lt;/li&gt;
&lt;li&gt;Plant voltage and frequency
&lt;/li&gt;
&lt;li&gt;Whether you want a dryer, crusher, chiller, or compressor included
&lt;/li&gt;
&lt;li&gt;Anything already fixed on site (for example an existing compressor), described in words rather than assumed to match every model
&lt;/li&gt;
&lt;li&gt;The three output figures you want quoted on your mold: kg/h, dry-cycle time, accepted pieces per hour
&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Payment and acceptance are separate topics, but they affect when utilities must be ready. Tongda’s factory acceptance covers machine inspection, running stability, mold confirmation, and product confirmation; payment is 30% advance and 70% before shipment. Ask for those checks to name the mold and product that were actually confirmed.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why engineers get this wrong
&lt;/h2&gt;

&lt;p&gt;Model names are easy to compare in a spreadsheet. Real output is not. Two machines in the same volume class can be set up with different heads, stations, and cooling. A plant that orders from the litre figure alone often discovers—after deposit—that the accepted pieces per hour on the actual mold do not match the schedule. The fix is boring and effective: demand the three figures on your mold, sketch the machine count with blanks until those figures arrive, and lock voltage, air, and cooling before the crate ships.&lt;/p&gt;

&lt;p&gt;For a quotation against your drawing, use: &lt;a href="https://www.tongdamachine.net/contact/" rel="noopener noreferrer"&gt;Contact Tongda&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>manufacturing</category>
      <category>engineering</category>
      <category>plastics</category>
      <category>packaging</category>
    </item>
    <item>
      <title>Panels, Sets and Shots: A Small Python Model for Blow Molded Table-Top Capacity</title>
      <dc:creator>Tongda Machinery</dc:creator>
      <pubDate>Fri, 30 Oct 2026 07:50:12 +0000</pubDate>
      <link>https://dev.to/tongdamachinery/panels-sets-and-shots-a-small-python-model-for-blow-molded-table-top-capacity-53l</link>
      <guid>https://dev.to/tongdamachinery/panels-sets-and-shots-a-small-python-model-for-blow-molded-table-top-capacity-53l</guid>
      <description>&lt;p&gt;Folding tables and camping tables often use a hollow, blow molded HDPE top. When a furniture maker plans one of these lines, the first spreadsheet usually takes a cycle time from a machine brochure, multiplies it by hours in a year, and calls the result "capacity". That number is almost always too high, and it is sometimes counted in the wrong unit.&lt;/p&gt;

&lt;p&gt;This post turns the planning logic into a few lines of Python. The reference case comes from a table-top article published by Suzhou Tongda Machinery, a Zhangjiagang-based extrusion blow molding machine maker. Their article on &lt;a href="https://www.tongdamachine.net/news/blow-molded-table-top-flatness/" rel="noopener noreferrer"&gt;controlling flatness on the TDW-250A&lt;/a&gt; covers the process side (wall distribution, cooling and support). Here we focus on the arithmetic.&lt;/p&gt;

&lt;h2&gt;
  
  
  The reported inputs
&lt;/h2&gt;

&lt;p&gt;The source reports these values for the TDW-250A table-top project:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Item&lt;/th&gt;
&lt;th&gt;Reported value&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Clamping force&lt;/td&gt;
&lt;td&gt;1,200 kN&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Accumulator&lt;/td&gt;
&lt;td&gt;28 L&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Maximum shot rate&lt;/td&gt;
&lt;td&gt;1.6 kg/s&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cycle&lt;/td&gt;
&lt;td&gt;120 s for one two-panel set&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Set weight&lt;/td&gt;
&lt;td&gt;about 5.8 kg in total&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Two things are worth noting before writing any code. First, the cycle produces a &lt;em&gt;set&lt;/em&gt; of two panels. A quote that says "30 per hour" could mean sets or panels, and the difference is a factor of two. Second, the maximum shot rate describes how fast the accumulator can discharge. It is not an output figure for finished tops.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 1: theoretical output, with the unit in the name
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;theoretical_output&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;cycle_s&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="nb"&gt;float&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;panels_per_cycle&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="nb"&gt;int&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;hours&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="nb"&gt;float&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="mf"&gt;1.0&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="n"&gt;cycles&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;hours&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="mi"&gt;3600&lt;/span&gt; &lt;span class="o"&gt;/&lt;/span&gt; &lt;span class="n"&gt;cycle_s&lt;/span&gt;
    &lt;span class="k"&gt;return&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="s"&gt;sets&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;cycles&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="s"&gt;panels&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;cycles&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="n"&gt;panels_per_cycle&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="nf"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nf"&gt;theoretical_output&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;120&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;2&lt;/span&gt;&lt;span class="p"&gt;))&lt;/span&gt;        &lt;span class="c1"&gt;# {'sets': 30.0, 'panels': 60.0}
&lt;/span&gt;&lt;span class="nf"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nf"&gt;theoretical_output&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;120&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;2&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;24&lt;/span&gt;&lt;span class="p"&gt;))&lt;/span&gt;    &lt;span class="c1"&gt;# {'sets': 720.0, 'panels': 1440.0}
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Returning both units from the same function makes it hard to mix them up later. These are still theoretical numbers: 24 uninterrupted hours with no rejects.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 2: accepted output
&lt;/h2&gt;

&lt;p&gt;Real lines stop for mold changes, color changes, material issues and maintenance, and some panels fail inspection. Model those as explicit inputs, not as a single "efficiency" fudge factor that nobody can audit.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;accepted_output&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;cycle_s&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;panels_per_cycle&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;planned_hours&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
                    &lt;span class="n"&gt;downtime_hours&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;reject_rate&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="n"&gt;run_hours&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;planned_hours&lt;/span&gt; &lt;span class="o"&gt;-&lt;/span&gt; &lt;span class="n"&gt;downtime_hours&lt;/span&gt;
    &lt;span class="n"&gt;t&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nf"&gt;theoretical_output&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;cycle_s&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;panels_per_cycle&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;run_hours&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
    &lt;span class="n"&gt;good_panels&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;t&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="s"&gt;panels&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;1&lt;/span&gt; &lt;span class="o"&gt;-&lt;/span&gt; &lt;span class="n"&gt;reject_rate&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
    &lt;span class="k"&gt;return&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="s"&gt;run_hours&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;run_hours&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
            &lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="s"&gt;good_panels&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;good_panels&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
            &lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="s"&gt;good_sets&lt;/span&gt;&lt;span class="sh"&gt;"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;good_panels&lt;/span&gt; &lt;span class="o"&gt;/&lt;/span&gt; &lt;span class="n"&gt;panels_per_cycle&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Plug in &lt;em&gt;your own&lt;/em&gt; plant's planned hours, recorded downtime and reject history. The function will not invent them for you, and that is intentional. If a supplier gives you an accepted-output figure that already excludes stoppages, do not subtract them twice.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 3: a sanity check on resin throughput
&lt;/h2&gt;

&lt;p&gt;The extruder has to plasticize the resin for every shot. A quick check compares the material demanded per hour with what the extrusion system is rated for:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;resin_demand_kg_per_h&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;set_weight_kg&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;cycle_s&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;flash_kg_per_set&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="mf"&gt;0.0&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="nf"&gt;return &lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;set_weight_kg&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="n"&gt;flash_kg_per_set&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="mi"&gt;3600&lt;/span&gt; &lt;span class="o"&gt;/&lt;/span&gt; &lt;span class="n"&gt;cycle_s&lt;/span&gt;

&lt;span class="nf"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nf"&gt;resin_demand_kg_per_h&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mf"&gt;5.8&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;120&lt;/span&gt;&lt;span class="p"&gt;))&lt;/span&gt;  &lt;span class="c1"&gt;# 174.0 kg/h before flash
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Flash weight is left as a parameter because it depends on the tool and trimming design. If the demand is close to the extruder's rated output, the cycle in the brochure may not be achievable with your part, and that is worth raising early.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 4: keep flatness out of the capacity model, but not out of the plan
&lt;/h2&gt;

&lt;p&gt;A panel that warps is a reject, so flatness affects &lt;code&gt;reject_rate&lt;/code&gt;. However, flatness is a separate acceptance criterion that needs its own definition: which surface, which supports, what temperature and how long after demolding. The source article makes the point that "perfectly flat" is not a production criterion without those conditions. Put the measurement method in the trial agreement, then let the measured reject rate feed the model.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to ask the equipment supplier for
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Cycle definition: what is counted, and is it a set or a single panel?&lt;/li&gt;
&lt;li&gt;Trial duration and the accepted count at a stated line boundary&lt;/li&gt;
&lt;li&gt;Part and flash weight from the trial, not from the drawing&lt;/li&gt;
&lt;li&gt;Mold envelope and machine revision used in the trial&lt;/li&gt;
&lt;li&gt;Take-out, trimming and any cooling fixtures included in scope&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Tongda lists its large-part machines, such as the &lt;a href="https://www.tongdamachine.net/tdw-350f/" rel="noopener noreferrer"&gt;TDW-350F for pallets and table panels&lt;/a&gt;, alongside the rest of its &lt;a href="https://www.tongdamachine.net/blow-molding-machine/" rel="noopener noreferrer"&gt;blow molding machines&lt;/a&gt; from 2 L to 1200 L. If you want to see the source data, the company's official site is &lt;a href="https://www.tongdamachine.net/" rel="noopener noreferrer"&gt;tongdamachine.net&lt;/a&gt;.&lt;/p&gt;




&lt;p&gt;&lt;em&gt;Written by Eileen Yu, Vice General Manager of Suzhou Tongda Machinery, with 20 years in the blow molding industry. Tongda, founded in 1999, makes hydraulic, servo and fully electric extrusion blow molding machines and has delivered more than 10,000 machines to more than 100 countries.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>manufacturing</category>
      <category>python</category>
      <category>planning</category>
      <category>blowmolding</category>
    </item>
    <item>
      <title>Modeling All-Electric Blow Molding Payback in a Few Lines of Python</title>
      <dc:creator>Tongda Machinery</dc:creator>
      <pubDate>Wed, 28 Oct 2026 03:01:31 +0000</pubDate>
      <link>https://dev.to/tongdamachinery/modeling-all-electric-blow-molding-payback-in-a-few-lines-of-python-2na8</link>
      <guid>https://dev.to/tongdamachinery/modeling-all-electric-blow-molding-payback-in-a-few-lines-of-python-2na8</guid>
      <description>&lt;p&gt;Plant engineers are often asked a deceptively simple question: "If we buy the all-electric blow molding machine instead of the hydraulic or hybrid one, when do we get the extra money back?" The arithmetic is easy. The hard part is feeding it inputs that describe the same product, measured the same way. This post turns that problem into a small, reusable model you can drop into a notebook or spreadsheet.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fxbui9bnt191zors6rdmu.jpg" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fxbui9bnt191zors6rdmu.jpg" alt="Extrusion blow molding machine" width="800" height="1097"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;The logic follows a payback guide published by Suzhou Tongda Machinery, an extrusion blow molding machine maker founded in 1999 in Zhangjiagang that supplies customers in more than 100 countries. Their article on &lt;a href="https://www.tongdamachine.net/news/all-electric-blow-molding-roi/" rel="noopener noreferrer"&gt;calculating all-electric blow molding ROI&lt;/a&gt; explains the reasoning in plant terms; here we restate it as code and a list of input rules.&lt;/p&gt;

&lt;h2&gt;
  
  
  The core idea
&lt;/h2&gt;

&lt;p&gt;Simple payback is just:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;payback_years = extra_installed_investment / annual_net_operating_savings
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Everything interesting lives in the denominator, and every term in it should be normalised to &lt;strong&gt;accepted bottles&lt;/strong&gt;, the ones that pass inspection.&lt;/p&gt;

&lt;h2&gt;
  
  
  Input rules before you write any code
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Same product on both sides.&lt;/strong&gt; Same bottle drawing, resin, target weight, mold, annual accepted quantity and schedule.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Metered energy, not nameplate power.&lt;/strong&gt; Use kWh per 1,000 accepted bottles from comparable runs. Power ratings are for sizing the electrical supply.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;State the meter boundary.&lt;/strong&gt; Machine only, or machine plus chillers, compressors, granulators and inspection? A smaller reading on the machine meter may not translate to the whole line.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Keep weight tolerance and average weight apart.&lt;/strong&gt; A tighter tolerance does not reduce average resin use unless a lower mean is actually demonstrated.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Count each improvement once.&lt;/strong&gt; If annual accepted output already reflects rejects and stops, do not also book a separate downtime saving for the same effect.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  A minimal model
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;energy_saving&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;base_kwh_per_k&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;new_kwh_per_k&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;bottles&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;price_per_kwh&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="c1"&gt;# kWh figures are per 1,000 accepted bottles
&lt;/span&gt;    &lt;span class="n"&gt;delta_kwh&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;base_kwh_per_k&lt;/span&gt; &lt;span class="o"&gt;-&lt;/span&gt; &lt;span class="n"&gt;new_kwh_per_k&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="n"&gt;bottles&lt;/span&gt; &lt;span class="o"&gt;/&lt;/span&gt; &lt;span class="mi"&gt;1000&lt;/span&gt;
    &lt;span class="k"&gt;return&lt;/span&gt; &lt;span class="n"&gt;delta_kwh&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="n"&gt;price_per_kwh&lt;/span&gt;

&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;resin_saving_tonnes&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;mean_reduction_g&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;bottles&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="c1"&gt;# only use a measured reduction in MEAN weight
&lt;/span&gt;    &lt;span class="k"&gt;return&lt;/span&gt; &lt;span class="n"&gt;mean_reduction_g&lt;/span&gt; &lt;span class="o"&gt;*&lt;/span&gt; &lt;span class="n"&gt;bottles&lt;/span&gt; &lt;span class="o"&gt;/&lt;/span&gt; &lt;span class="mi"&gt;1_000_000&lt;/span&gt;

&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;simple_payback&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;extra_investment&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;annual_net_saving&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="k"&gt;return&lt;/span&gt; &lt;span class="n"&gt;extra_investment&lt;/span&gt; &lt;span class="o"&gt;/&lt;/span&gt; &lt;span class="n"&gt;annual_net_saving&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Demand charges, tax, financing and residual value are deliberately left out. Simple payback is a screening number, not a full investment appraisal; it ignores the time value of money.&lt;/p&gt;

&lt;h2&gt;
  
  
  Plugging in the published example
&lt;/h2&gt;

&lt;p&gt;Tongda's guide uses a hypothetical set of inputs to show the arithmetic: 8,000,000 accepted bottles a year, 25 versus 18 kWh per 1,000 bottles, USD 0.15 per kWh, USD 1,600 of other net savings and USD 30,000 of extra investment. Running those through the functions:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="n"&gt;e&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nf"&gt;energy_saving&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;25&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;18&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;8_000_000&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mf"&gt;0.15&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;   &lt;span class="c1"&gt;# 8400.0
&lt;/span&gt;&lt;span class="n"&gt;net&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;e&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="mi"&gt;1600&lt;/span&gt;                               &lt;span class="c1"&gt;# 10000.0
&lt;/span&gt;&lt;span class="nf"&gt;simple_payback&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;30_000&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;net&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;                  &lt;span class="c1"&gt;# 3.0 years
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The guide is explicit that these are illustrative inputs, not a quotation and not a measured result from any specific machine. That matters: the "three years" is a property of the inputs, not of all-electric drives in general.&lt;/p&gt;

&lt;h2&gt;
  
  
  Add a sensitivity loop
&lt;/h2&gt;

&lt;p&gt;Utilisation is the lever most people underestimate. Hold everything else constant and vary volume:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;for&lt;/span&gt; &lt;span class="n"&gt;bottles&lt;/span&gt; &lt;span class="ow"&gt;in&lt;/span&gt; &lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;8_000_000&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;6_000_000&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;4_000_000&lt;/span&gt;&lt;span class="p"&gt;):&lt;/span&gt;
    &lt;span class="n"&gt;net&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nf"&gt;energy_saving&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;25&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;18&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;bottles&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mf"&gt;0.15&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="mi"&gt;1600&lt;/span&gt;
    &lt;span class="nf"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;bottles&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nf"&gt;round&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;net&lt;/span&gt;&lt;span class="p"&gt;),&lt;/span&gt; &lt;span class="nf"&gt;round&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nf"&gt;simple_payback&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="mi"&gt;30_000&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;net&lt;/span&gt;&lt;span class="p"&gt;),&lt;/span&gt; &lt;span class="mi"&gt;2&lt;/span&gt;&lt;span class="p"&gt;))&lt;/span&gt;
&lt;span class="c1"&gt;# 8000000 10000 3.0
# 6000000 7900 3.8
# 4000000 5800 5.17
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;These match the sensitivity table in the original guide. In a real model, revisit which maintenance and operating costs scale with utilisation instead of holding them fixed.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fdbvrkv8py5e6sabksvzw.jpg" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fdbvrkv8py5e6sabksvzw.jpg" alt="Blow molded bottles" width="800" height="533"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Resin as a separate scenario
&lt;/h2&gt;

&lt;p&gt;The guide also shows that, at 8 million bottles, a demonstrated 1 g reduction in mean bottle weight equals 8 tonnes of bottle material per year (&lt;code&gt;resin_saving_tonnes(1, 8_000_000)&lt;/code&gt;). Keep this out of the energy figure unless you have measured it, and remember that flash, rejects, regrind and losses all affect actual resin purchasing.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to ask a supplier for
&lt;/h2&gt;

&lt;p&gt;To fill the model with real numbers, request accepted output, metered kWh with the boundary stated, average bottle weight and its variation, reject rates, utility conditions, annual operating hours and the full installed-cost scope. Keep theoretical ratings in a separate column from test results.&lt;/p&gt;

&lt;p&gt;If you are evaluating equipment, Tongda's &lt;a href="https://www.tongdamachine.net/fully-electric-blow-molding-machine/" rel="noopener noreferrer"&gt;fully electric blow molding machine&lt;/a&gt; series page lists the current HSE models, and the company can be reached through its official website, &lt;a href="https://www.tongdamachine.net/" rel="noopener noreferrer"&gt;tongdamachine.net&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>manufacturing</category>
      <category>python</category>
      <category>energy</category>
      <category>blowmolding</category>
    </item>
    <item>
      <title>Inside the Control Loop of an Extrusion Blow Molding Machine</title>
      <dc:creator>Tongda Machinery</dc:creator>
      <pubDate>Tue, 27 Oct 2026 15:56:00 +0000</pubDate>
      <link>https://dev.to/tongdamachinery/inside-the-control-loop-of-an-extrusion-blow-molding-machine-9b5</link>
      <guid>https://dev.to/tongdamachinery/inside-the-control-loop-of-an-extrusion-blow-molding-machine-9b5</guid>
      <description>&lt;p&gt;Extrusion blow molding (EBM) looks simple: extrude a tube of molten HDPE (the parison), close a mould around it, blow, cool, eject. Under the guards, a modern machine is a set of coupled control problems. Here are four: parison wall thickness control, servo hydraulics, PLC/HMI integration, and measuring energy. Machine examples come from the published documentation of Suzhou Tongda Machinery, a Chinese EBM manufacturer.&lt;/p&gt;

&lt;h2&gt;
  
  
  1. Parison control: a fast position servo
&lt;/h2&gt;

&lt;p&gt;A hanging parison stretches under its own weight, and a blown container thins out most where it expands the most. Without control, the only way to hit the minimum wall at the thinnest spot is to make everything else thicker. Parison programming fixes this by changing the die gap while the parison is extruded.&lt;/p&gt;

&lt;p&gt;The usual architecture is a closed position loop:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;a wall thickness controller holds the programmed profile, a curve of die gap against parison length;&lt;/li&gt;
&lt;li&gt;it sends a voltage or current command to an electro-hydraulic servo valve;&lt;/li&gt;
&lt;li&gt;the valve drives an actuator that moves the die core up and down, changing the annular gap;&lt;/li&gt;
&lt;li&gt;a linear position transducer reports the actual gap back to the controller.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The timing is demanding. Tongda's technical note works through a simple 10-point profile on a continuous machine: if the parison is extruded in about 5 s, each segment lasts roughly 0.5 s, and the loop's step response should be 1/5 to 1/10 of that, around 0.05 to 0.1 s, while still tracking the set point accurately. Because a 10-point curve is coarse, the controller interpolates between points to keep the profile smooth. Commercial controllers offer 30 or 100 points for finer profiles.&lt;/p&gt;

&lt;p&gt;The payoff is material. The same note cites a wide-mouth barrel that weighed 3.5 kg without parison control and met the same minimum wall at 2.4 kg with it, a raw material saving of 20 to 30%.&lt;/p&gt;

&lt;h2&gt;
  
  
  2. Servo hydraulics: power on demand
&lt;/h2&gt;

&lt;p&gt;A conventional machine runs a fixed pump continuously and throttles the surplus into heat. A servo hydraulic machine drives the pump with a servo motor that runs only as fast as each movement needs: less energy, cooler oil, faster and more repeatable movements.&lt;/p&gt;

&lt;p&gt;The benefit scales with the machine. Tongda builds its larger machines, from 50 L drums up to 1,200 L IBC tanks, on servo drives. Mixed architectures are common too: on its double-station HS II-30L, the blow pin, die head lift and platform lift are electric, while the main movements use a hydraulic servo system with proportional valve control.&lt;/p&gt;

&lt;h2&gt;
  
  
  3. PLC and HMI: one screen for the whole line
&lt;/h2&gt;

&lt;p&gt;The machine PLC sequences the cycle: extrusion, mould close, cut, carriage move, blow, exhaust, open, eject. On Tongda's large accumulator machines, a Mitsubishi PLC with a Chinese/English touch screen handles process setting, changes, retrieval, monitoring and fault diagnosis.&lt;/p&gt;

&lt;p&gt;The more interesting trend is integration beyond the machine. On the fully electric &lt;a href="https://www.tongdamachine.net/hse-12l-blow-molding-machine/" rel="noopener noreferrer"&gt;HSE-12L&lt;/a&gt;, a 15-inch touch screen monitors the main machine and each auxiliary unit, the main and auxiliary machines are linked over Modbus/TCP, servo wall thickness control is integrated, and an IoT function lets production status be viewed from the cloud. The control system also provides early warning alerts and stores production data for analysis.&lt;/p&gt;

&lt;p&gt;From a software point of view this is familiar ground: a register map per auxiliary (chiller, loader, mould temperature controller), polling or event-driven reads, alarm aggregation on the HMI, and time-series storage upstream for OEE and maintenance.&lt;/p&gt;

&lt;h2&gt;
  
  
  4. Measuring energy honestly
&lt;/h2&gt;

&lt;p&gt;Energy claims are easy to make and hard to compare. A sound method uses general power-quality practice:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;measure at the machine's main supply with a three-phase power analyser, ideally logging heaters and drives separately;&lt;/li&gt;
&lt;li&gt;record over a stable production window, not during start-up;&lt;/li&gt;
&lt;li&gt;normalise to output: kWh per kg of resin processed or kWh per good part;&lt;/li&gt;
&lt;li&gt;compare machines on the same part weight, cycle time and resin.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Tongda's own figures follow this logic. Its large-machine plasticizing system quotes an average of 3.97 kg plasticized per kWh. Its buying advice is equally practical: ask for the hourly energy consumption on your own product.&lt;/p&gt;

&lt;p&gt;For the full background on profile control, see Tongda's note on &lt;a href="https://www.tongdamachine.net/implementing-wall-thickness-control-in-extrusion-blow-molding-machines/" rel="noopener noreferrer"&gt;implementing wall thickness control&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>manufacturing</category>
      <category>automation</category>
      <category>engineering</category>
      <category>plastics</category>
    </item>
  </channel>
</rss>
