Need a Survey?
020 3931 5759
June 12, 2018

IDS Stream EM Review

GSSI Structure Scan Mini XT review

A lot of claims have been made regarding the IDS (Leica) Stream EM. As a former owner of a business (GPR Pro Ltd – now sold and of no affiliation) operating one of these pieces of equipment I (Kevin Banks –previously the managing director of GPR Pro and sometime prior to that a direct employee of IDS) want to take this opportunity to clarify what the Stream EM is, and what its capabilities are, and what they are not.

Get A Quote
High Quality Surveys
Satisfaction Guaranteed
Safety First in Every Project
Our Clients

We Care About Our Clients' Experiences

View More
  • Luis Brooks


    I cannot recommend KB GPR highly enough. The service they provide is second ...

  • Archer Baxter


    This company is superb. My situation was a bit of a tricky one but they took...

  • Jay Carey


    The service I received from KB GPR Surveys was amazing. They were so efficie...

Need More Info?

Talk To Our Experts Now
020 3931 5759
Get A Quote

Talk To Us

Simply complete the form and we’ll be in touch.

    The IDS Stream EM is a 38 channel, towed, high density Ground Penetrating Radar (GPR) array which is designed to allow the collection of GPR data in one direction only, whilst reproducing the equivalent of a full GPR survey in a traditional, orthogonal grid (requiring the GPR to be pushed in two directions perpendicular to one another). This data is collected using either GPS or Total Station for positioning and is designed for connection to a vehicle. This makes it appropriate for surveys of roads, but also a reasonably efficient technology to survey large open spaces.

    What is the Stream EM?

    The purpose of the design is to allow the collection of B-Scans in both directions at the same time with only one pass of the GPR, thus increasing efficiency, however, the generally accepted side effect is that by gathering dense blocks of GPR data, you will also be able to generate high quality images of the below ground environment and that this will increase the quality of your processing and make it easier and more efficient.

    The advantages of using this type of system are:

    • No need to mark out survey grids (time saving)
    • No need to travel in two directions to perform the survey (time saving)
    • Less requirement for work in-road working (improved safety)
    • No requirement for fixed traffic management (reduced cost and less local disruption)
    • High density of scanning (confidence in results and improved imaging)

    The Stream EM is a complicated machine, designed to be constructed at the beginning of a shift, used for the survey, and then deconstructed and transported away from site at the end of the shift. The construction takes a typical team who know what they are doing about 40 minutes (assuming they are moving at normal speed but not racing against the clock), if staff are less diligent, it can take over an hour.

    Building a Stream EM requires two men and involves manual handling of heavy parts. It is not difficult, but it’s certainly not a ‘convenient’ process. It can be infuriating and sometimes requires use of brute strength. Occasional assembly doesn’t offend anyone, but repeated assembly and dismantling shift after shift will wear down even the most enthusiastic staff.

    Assembly of the Stream EM, towed high density GPR array

    To be transported, the Stream EM requires a large vehicle. GPR Pro operated a long wheel base, high roof van which was custom fitted out to accommodate all the components of the Stream securely and safely. The Stream EM could be packed in a smaller vehicle but there would be less opportunity to secure the equipment during transit – at GPR Pro, we felt it justified the large van. We also hated the large van for all the inconveniences which came with it.

    After assembly, either a Total Station or a GPS should be set up for use with the Stream and then data is collected by ‘driving around’ until the entire survey area is covered. Of course, a structured method for data collection is advisable, ‘driving around’ does not produce good survey results.

    Operationally, use of the Stream is very straightforward. If using GPS there is little more to it – if using a Total Station, empathies is required for the topo elements to maximise efficiency and minimise difficulties during processing.

    There are of course issues with the Stream, I won’t dwell unduly on the multitudes of problems we had to work through because many of them have been resolved by IDS at the time of writing, but even so, it’s worth mentioning a few of them.

    There are some bugs within the software, I would say that at this stage they are not a major issue for the equipment. (At the time we purchased ours it was not fit for purpose – if fact it was so bad as to be virtually unusable.)

    The equipment is prone to breakages, this is due to the complexity of the design and the number of parts:

    • Screws will fall out whilst you are transporting the Stream in your vehicle, if you don’t detect and tighten them parts may fall off whilst you are surveying.
    • Straps are likely to snap without warning. On one memorable shift two straps snapped under load and the entire central part of the trolley flipped upside down and was dragged several meters in this position before the driver stopped. This caused significant damage to three antennas and a complete write-off of another, we also snapped several cables.
    • Due to the large area coverage, the soft plastic skid trays can be worn through in a single shift – if this isn’t detected quickly in time you will wear through the bottom of your plastic antenna cases and incur expensive repairs.
    • You can expect to blow at least one fuse per shift, however without the electric winch you will find the machine unusable.
    • The antennas are not waterproof, 6 out of 8 of our antenna boxes (containing multiple antennas inside) broke due to water damage in the first 12 months.

    The list goes on and on, mostly minor issues that should not happen, but all significant causes of damage and delay. None of these things should be possible on a well-designed piece of equipment in this price range but all of them need to be prepared for so that a shift can continue smoothly without stopping the job.

    A two-man team can easily operate the Stream. If using Total Station, a third pair of hands is helpful, but not essential.

    Although marketed as not requiring TM an escort vehicle will be required. In the UK, the Stream EM trailer is not road legal and wider than the maximum allowable towing width. It is illegal to operate without an escort, therefor you are also operating without insurance if you attempt to use it without an escort. You’ll find that other road users neither accommodating or cooperative towards a slow-moving obstruction on the road.

    The trailer design has protruding parts, you’ll need to judge how much of the road you are going to cover (kerb to kerb, minus the margin of safety you allow for not accidentally hitting the kerb). In addition to this safety distance, there will be a further 50cm between the furthest protruding edge of the Stream and the first antenna dipole. It’s virtually impossible to safely scan closer than 70cm to a kerb edge of a bending road or obstruction.

    Data collection itself is quite straightforward, I won’t publish a full set of recommendations, but basically you want to perform parallel scans with as few gaps as possible and as minimal overlap as possible. Both gaps and overlaps will result in a reduction of data quality.

    The software for data collection is quite acceptable and there are no issues to use it, at this stage the obvious bugs have been worked out. It is not designed to be customised and in fact IDS do everything they can to prevent meaningful customisation. We were able to change settings on our own Stream and did so without IDS blessing, it was a necessary requirement because of the number of breakages and equipment failures we had to deal with.

    Antennas and other parts used to break regularly and we also had to allow shift time in the schedule for dealing with problems.

    Even so, the Stream was very efficient and noticeably more so than traditional survey methods.

    On good ground the data from the Stream was exceptional (as it undoubtedly would have been using any GPR), please click the link to view a short video showing an area of Gatwick airport taxiway. This beautiful image leaves little doubt as to the usefulness of a towed high-density array, however, we also demonstrated that we could achieve equally beautiful results on the same ground using a traditional GPR in a grid. On a direct comparison basis, it leaves a lot of questions as to the need for such an expensive and complex machine: in practice, we preferred the grid.

    **space holder for link **

    On bad ground, you could see next to nothing, it is then necessary to work through the data in the traditional way. Please click the link for a video showing some GPR data we collected in which the Stream images were useless. We would have had much more confidence if we had had a GPR grid to process, including confidence that if the results were poor, it was because of the ground and not the equipment.

    **space holder for link **

    On any instance with positioning problems the Stream data would not be good, fortunately IDS have resolved most of the time lag issues in their software and this is one thing that now works quite well (it didn’t at the time of purchase).

    Managing and processing Stream data requires organisation, it’s fair to say that a lot of companies can do it, but not a lot of companies do it well. I do not intend to publish our recommended post processing procedure. IDS GRED HD3 post processing software is quite annoying, but it is an efficient means to deal with Stream data and to process utility survey data in general and is recommended.

    You load the GPR data into the software, run a data ‘conversion’ to create .dt files which the software can read, the positioning is automatically read from the GPS/Total Station/Grid information recorded on-site, and processing is completed for you. The user must perform data interpretation which means plotting the positions of pipes and features within the data. Do not attempt to use non-standard settings on GPR and then use this software without testing it thoroughly first, it is not a flexible system.

    The data merge between the front and rear arrays doesn’t work well, so half of the 30 VV channels are of little practical use. The software works faster and reconstructed B-scans look superior if only half of the data is used. IDS had removed the facility to directly switch off half of the array, but it is still possible by deselecting individual channels.

    This means that the effective spacing between samples of the reconstructed A-scans would not be 6cm as advertised but 12cm.

    12cm does not meet the PAS128 specification for a high-density array (although the validity of PAS128’s 10cm limit with no reference to frequency or wavelength is questionable). Of course, the machine itself does meet the specification.

    The typical distance between A-Scan samples for a pushed GPR with a survey wheel is usually between 1.5 and 2.5cm. 6cm is the best that can be achieved for the reconstructed B-scans produced by the Stream and that is at least 2.5 times lower than a typical GPR gridded survey, in practice 12cm was more common and that means it’s at least five times less than a system pushed in a grid.

    This is using an array of 200MHz antennas which provide relatively low-resolution data. The idea of ‘X-raying’ the ground using a Stream is complete nonsense: there will be targets that the Stream EM simply cannot see.

    Overall the high-density array is a good idea and with good management it does work and can generate good results.

    To review against the original points:

    • No need to mark out survey grids (time saving) – true
    • No need to travel in two directions to perform the survey (time saving) – true

    The Stream EM is more efficient when is used to survey large areas and roads using GPS, or in busy city centres using Total Station for positioning. For inner city areas, areas with restricted manoeuvrability, small areas, and roads without GPS coverage, with the time lost to turn around between scans, assembly and disassembly time, and broken radar down-time, a GPR survey in an orthogonal grid will be more efficient.

    • No requirement for in-road working (improved safety) – true as long as it keeps working, in practice staff had to jump out of the van and adjust cables or restart the equipment from time to time.
    • No requirement for fixed traffic management (reduced cost and less local disruption) – true but an escort vehicle is required

    Overall, use of the Stream is safer than working in the road.

    • High density of scanning (confidence in results and improved imaging) – whilst the scanning density is high, the edges of roads and areas around obstructions will be missed. The image quality is very variable, sometimes images will be good and at other times poor.

    Confidence in results is not better than a traditional gridded survey, particularly given the resolution of the system. In fact, GPR surveys using the orthogonal grid method produce more reliable results.

    To customers of Stream EM surveys: The Stream EM is suitable for efficient coverage of large areas or surveys of roads for a reasonable price. Quality can be acceptable but will be lower than a gridded GPR survey: it stands to reason that the m2 cost for a Stream EM survey should be less to match the increased efficiency on-site and lower quality outputs. Don’t pay a premium for a survey with a Stream EM.

    To GPR surveyors: At GPR Pro we disliked our Stream EM and as the owner of different GPR business I wouldn’t recommend one if asked and I would avoid buying one again. I do see a place for the towed GPR system – but should I ever require one in the future I would build it myself before I would purchase a Stream EM. I have not investigated alternatives by different manufacturers although in time I will do so.

    Personal note: I have tried to keep these comments mostly neutral however I should clarify that we encountered many more problems with the Stream EM and support from IDS than have been mentioned here. I’m proud to report that we never failed a job, but that was a reflection on our ability to work around problems and not on the quality of the equipment or support that we were provided.

    Stream data quality: The data quality is okay if the process is well managed, but it’s not amazing and it’s not as good as a traditional GPR survey using dense orthogonal grids. If it’s not well managed, outputs will be poor.

    Some of the issues we encountered have since been resolved and the software processing as it stands at the time of writing is substantially different from terrible (not an overstatement) experiences we had with earlier versions of the GRED post processing and One Vision data collection software’s. The system does/can work.

    Nevertheless, based on our experiences with the Stream and dealing with IDS, I would advise against the purchase of a Stream EM.

    kb gpr surveys expert
    Talk to us today

    Ready To Get Help From 
    The Experts?

    Get in touch to see how we can help you today!

    Related Posts

    Ready to find out how we can help?

    Contact KBGPR Surveys today for a quick response – we are always happy to offer advice.
    2022  © Copyright KB GPR Surveys Ltd.