Questions |  | Answers |
|
|
| |
| Why don’t I get the maximum throughput stated for my system? | The 4100, 4500 and 4600 have a maximum mechanical throughput of 1200 DPH.
The 3100, 3500 and 3600 have a maximum mechanical throughput of 700 DPH.
The 4700 has a maximum mechanical throughput of 1400 DPH.
The 3700MK2 has a maximum mechanical throughput of 950 DPH.
The quoted mechanical throughput for any system from any vendor is maximum throughput achieved under ideal conditions. These assumed conditions include:
1. Short programming times where the handler always has a part to pick or place.
2. The input/output and programming sites are in close proximity to each other.
3. No failures occur.
4. Using the maximum number of sockets.
It is possible to achieve the maximum stated throughput on your system for programming times less than 10 seconds depending on your input/output media. As programming times increase, throughput starts to decrease. The handler is no longer operating at maximum speed because it is periodically waiting for parts to finish programming.
If you are laser or label marking and the combined time to mark and program exceeds 10s you will get less throughput as well. Remember to use the combination of programming and marking time when benchmarking throughput.
NOTE: Small lot sizes may not run long enough to achieve maximum throughput |
|
|
| Why can’t I get 9,600 devices programmed in one 8-hour shift on my 4600 APS? | Although the 4600 is capable of 1200 DPH, it is not possible to sustain this over an entire shift. Throughput over an entire shift can be affected by a variety of factors such as, changeover time, reloading feeders, job setup, and programming yields.
Operator efficiency can have a significant affect on throughput numbers. Systems may sit idle during a programming job waiting for blank parts or because some other event has occurred requiring attention such as breaks or other tasks the operator is occupied with at the time. Operators can help increase your throughput by staying alert to the needs of the system while a job is in progress. Also, operator skill and knowledge can greatly affect throughput. The amount of time it takes to set up a job, troubleshoot or re-load a tray stacker or a tape feeder varies from person-to-person and from job-to-job.
NOTE: BP Micro’s quoted throughput times assume a non-stop operation. This can be achieved on small jobs that do not require a re-load, or for input/output via tubes. You can view your operating efficiencyin the Job Summary. |
|
|
| Why is the system not using all of the programming sites but I’m still not achieving maximum throughput? | While it is possible to achieve maximum throughput using fewer than the maximum number of programming sites, the programming times must be very short. For programming times greater than 10 seconds, you will need all sites to maximize throughput.
The handler will automatically optimize by itself, using the number of programming sites required for maximum throughput, and it will chose the sites that require the least amount of travel. However, it is best to start a job with all sockets in place and let the handler do the optimizing. If left up to an operator, they may not choose the correct number of sockets or place the sockets in a pattern that is most efficient.
The system does not necessarily achieve maximum throughput when it optimizes the number of sites used. It is optimizing for the maximum throughput achievable for that programming time and any additional travel time to extra sockets will degrade throughput, therefore not all programming sites will be used for every job. |
|
|
| What can I do to increase throughput ? | · Use the maximum number of programming sites. The sites closest to your input/output media will provide the most throughput. The farther the PNP head must travel, the lower the throughput. The Master Site is normally set to site 1, but this can be changed. For example, in some cases, setting the Master Site to site 7 can improve throughput.)
· Throughput falls when parts are rejected. Rejects can be caused by bad teaches, dirty sockets, or bad parts. To maximize throughput, re-teach sockets if they are producing a low yield. Frequent continuity errors are usually the result of a bad teach or dirty socket leads. Clean sockets prior to each use.
· Use FX modules for Flash with the 6th generation systems.
· Train operators to be alert and act immediately when system requires attention. The system can sit idle for a considerable amount of time while waiting for trays to be changed or a mis-pick to be cleared, if the operator is busy with some other task.
· Check that the mod codes for motor speed, vacuum delay and alignment method are set for optimal performance. Sometimes during system troubleshooting a technician may derate the motor speed or increase the vacuum delay and forget to return it to the original setting. |
|
|
| What makes an FX module faster than a standard module? | These socket modules have been engineered to take advantage of the 6th Generation special high-speed pin drivers to produce the fastest programming times on the market. These pin drivers must be connected directly to the data and address pins of the DUT.
Most Flash Memory parts will show a great improvement in programming times when using an FX module (up to 50% faster in some cases) vs. a standard module. As a general rule, use an FX module when programming Flash Memory parts. (FX modules only work on BP Micro’s 6th generation programmers.)
FX modules are not available for every device. Not all device types will benefit from these modules due to chip architecture and already short programming times. |
|
|
| Can you make these programming times any faster? | In some cases, our Device Support engineers can add further optimization to an algorithm to make the part program faster. This can be due to an updated specification or a new die revision. |
|
|
| How come my programming time benchmark does not match the one given by BP Micro? | When comparing programming times it is important to do an apples-to-apples comparison.
· Did you choose the same configuration options (i.e. Program + Verify Twice)?
· Did you use the same file?
· Did you use the same programmer model?
· Did you use the same socket module?
· Did you record the master site time or the concurrent (2nd part) time?
Benchmarks quoted by BP Microsystems represent the best estimate of the maximum throughput you can expect based on observed programming times, but BP Microsystems does not guarantee your results. The quoted throughput represents the maximum run rate of the equipment and does not factor in set up time, operator response time, downtime for maintenance, failing to use the equipment to it’s full potential, etc. In addition, many factors may affect the programming time and hence the maximum run rate, including but not limited to data pattern, programming options, changes in device algorithm or specifications, die revisions, changes resulting from repeated erasure and programming, and other factors. |
|
|
| What is the formula for calculating throughput? | BPMicro does not publish a formula because the relationship between programming time and throughput is non-linear. The handler software attempts to maximize throughput by selecting the next-closest site each time a part is programmed. Minor differences in programming times can cause changes in the “pattern” of sites used.
We cannot guarantee the programming times or throughput on your devices will match a benchmark done at our factory. There are many factors when programming a particular device that produce different programming times. Devices from different lots or even the same lot may have significantly different programming times due to the manufacturing processes. Different data patterns or the number of times a part has been programmed or erased can also produce different programming times. |
|
|
|
|
| Why is it important to compare Program + Verify times without Erase for Flash? | This is the standard convention for showcasing the best times for programming Flash Memory. Flash Memory parts erase themselves. In other words, the programmer is not controlling the Erase function and cannot decrease the time it takes. Because of this, Erase times on all programmers are the same.
To truly compare programming times for Flash Memory, look at the Program, Verify, or combined Program + Verify times. |
|
|