AMPIC – ARM multi-protocol interface converter

Having been the innovative market leader in the embedded and industrial industry for the last 2 decades, DSL has always had a precise idea of what the industry is missing – however rarely do we get to seal 3 gaps with the same plug!

Remote Sensors – the gap

Remote sensors to date generally fall into two categories. The first: Off the peg ‘solutions’ that are restrictively generic, thus under or over specified for the required task. This can be in the form of functional deficiency, perhaps not meeting the level of accuracy truly required, or the frequency of which it can be measured. Alternately it provides functionality overkill, whilst, yes, meeting the needs of the application; it’s plethora of extraneous functions, bells and whistles included stretches budgets unnecessarily at best, at worst, is simply profits down the proverbial drain.

The second: the ‘from scratch’ custom route. Whilst this will always get you exactly what you want, costs on occasions can be impossible to amortise, with a long time to market and doesn’t cover the eventuality that not all installations will need exactly the same functionality. The software controlling this will also need to be designed, potentially expensively, from scratch.

To summarize a typical remote sensor application, I’ll use the example of an Ethernet enabled temperature sensor, utilizing an on board temperature sensor or thermocouple.

Standalone control systems – the gap

The same historic industry issues that apply to Remote Sensors also apply to more intelligent set ups, where a local system is not only monitoring variables via sensors, but also controlling outputs based on those variables.

Steps have been made to modularize such setups but maintaining software support through potentially multiple manufacturers’ devices can be problematic at best. As above, undertaking a completely ground up design becomes costly and doesn’t permit what typically are varying installation environments.

As an example, I’d illustrate a refrigeration controller, measuring temperatures and intelligently making decisions based on input variables to control fans/compressors, using relays.

Industrial I/O – the gap

Your typical industrial I/O convertor, whilst a solid product again suffers the same untenable traits as the sensor based systems. Whilst all a system integrator is trying to do here is convert from one interface to another, rarely is it that simple.

These products tend to be produced in low volume, naturally at huge detriment to their cost effectiveness. In an attempt to reduce the number of products in a range (naturally, to increase the manufacturing quantity, to reduce unit cost) your typical industrial I/O converter becomes bloated with functionality that again, one rarely needs. Especially when dealing with ADC/DAC, or other less popular I/O, with expensive ICs – an over specified product quickly becomes a very expensive one.

Again, one could design what they need from scratch, though unit costs will be far more palatable, the NRE costs involved are unlikely to be. Unless you’re manufacturing in the millions this is unlikely to be a particularly attractive route either – and you have the same lack of modularization problems.

Our concept

Addressing all of these issues, DSL has worked tirelessly to both identify and produce a true half way house. One that achieves the advantages of both, whilst loses the disadvantages of each. Harnessing the power of the latest ARM microcontrollers, DSL’s true ‘eureka’ moment came as it realized it could kill an unprecedented 3birds with a single embedded stone.

The solution is a building block approach, made possible by our AMPIC – ARM Multiprotocol interface converter.  Using this development platform, a client is able to proof of concept any type of remote sensor, standalone sensor based control system or industrial I/O conversion requirement instantly. With a full, all-encompassing software suite provided ready to run from the box and the expert guidance of DSL’s dedicated support team one can prove the solution works before the end of the same day!

AMPIC block diagram

Having expediently satisfied this, the next simple stage is the splitting the metaphorical functional building blocks of the development board to retain just the functionality that is required.

Having long since undertaken and componentized the schematics, footprints, layouts and released and integrated the complex translation software, before our client base have even heard of our new product, DSL are three quarters through the design cycle before our client’s say go!

Merely spinning off a client specific production design defined by the I/O areas the client wishes to keep, is child’s play for DSL. They simply define what is needed (and what isn’t) and DSL will produce a ‘custom’ product ready for production within weeks!AMPIC software stack

Additionally, by modularizing the extracted functionality, varying installation scenarios can be covered without duplicating the excess costs of over functionality in our earlier example.


The advantages needn’t stop there. A truly custom solution looks the part, it lives in an enclosure specified rigidly by the customer, from shape, size to colour scheme – and this isn’t lost with DSL’s approach. We can produce the module either to your own rigid mechanical specification, or design, produce and test a customized enclosure delivering you a finished product.

Off the shelf hardware can only plug the gap for low quantities, where the odd few pounds will have little bearing. If the project is intended for any reasonable quantity, this soon mounts up to a significant amount that, especially in these budget conscious times, needs addressing.

Whilst specifically this product is new for DSL, the concept is not – Utilising our historical experience in x86 PC based custom design and the radical concept of System On Module design DSL introduced all those years ago, we can state with confidence that we always deliver what is required, when it’s required.