New DL350 Portable Scopecorder

[Yokogawa TM UK]

Hey Everyone,

Yokogawa launched the DL350 Scopecorder this week and we have all been run off our feet!

In simple terms, the DL350 is a portable and fully isolated, electro-mechanical-recording-scope.

The user can choose what type of test instrument it is with 18 different modules types that can measure, trigger and record a variety of different measurements. If applications change, then you can easily change the modules to suit. It can be used in the test environment and in the field too and the resistive touch screen offers noise immunity and clarity of the measurements being made.

Recording up to 50 days and an independent battery life of 3 hours.
Up to 32 channels, including temperature, strain, acceleration, frequency, logic, CAN, LIN, SENT in addition to Voltage and HV modules too.
Up to 16 bit vertical resolution with split graticules to maintain the signal integrity, also up to 100MS/s sampling rate too which suits a hoard of applications.
Record length of 100M per module - shared between active channels i.e. if only one channel in use then 100M Record Length, 2 channels = 50M etc.

In addition, features such as GPS, Math, FFT as well as actions on trigger and an external I/O allow for great flexibility for a wider range of users requirements.

Sounds like a lot of features which usually means a lot of options - the only options on the DL350 are:
A Vehicle Edition Firmware Upgrade for CAN, LIN and SENT Modules
Battery Pack

.....thats it! Everything else listed above except the modules is included in the chassis.

The DL350 chassis is priced separately from the input modules, the full specifications can be found on page 15 on the following datasheet: https://www.yokogawa.com/pdf/provide/E/GW/Bulletin/0000031758/0/BUDL350-01EN.pdf


I am curious to know what pricing you would put on this?

[nctnico]

First impression: it looks like a MicSig tablet scope. Secondly: include the CAN/LIN decoding decode options as standard. And where are the I2C, SPI and UART decoders? And why is the FFT only 2048 points? 1Mpts FFT is today's standard! You will get remarks about the resistive touch screen but I guess that has to do with operators wearing gloves? I'm not so sure that is a good choice because this scope is likely to be used in R&D environments and people don't wear gloves even if they work on a car or truck (been there).

Pricing? IIRC Keysight's high-end portable scope with isolated input and 200MHz bandwidth is somewhere around $3500 so I guess this scope has to be in the same ball park.

[Yokogawa TM UK]

I can see the similarities in appearance with the MicSig - although this is a different unit altogether. The scopecorder series is not solely an oscilloscope - we have the DLM series which offers a better fight in that market, including the I2C, SPI, UART and more.
2048 points on each FFT (two independent windows), is reasonable in the market.
If comparing to scopes though, some of the basic specs would be interesting to observe such as record length which (as I assume we all know) effects sampling rates over measurement time. Referring back to the mentioned MicSig with an advertised 2GS/s and record length of 90MPoints - this means that you only achieve 2GS/s if your measurement time is no greater than 45ms.

On the Resistive touch screen, it is more than about wearing gloves. Capacitive touchscreen can be subject to many different noise sources, both internal & external. Charger and display noise are two of the most common and problematic noise sources today. As devices get thinner and chargers get noisier, these challenges will only become tougher to manage. Even everyday items can generate noise that causes interference, such as radio signals, AC mains, even fluorescent lightings. In the presence of noise, the positions reported by capacitive touch systems become distorted impacting accuracy and reliability. 

Resistive touch screens however don't experience this as both of the layers face each other & are separated with a thin gap which creates an electrical resistance between both layers. Therefore the result is a stable display that can be relied upon.

Thanks for your feedback

[JPortici]

but at least it has SENT. only portable solution i know is a picoscope and a laptop, others are all benchtop scopes.
Everything in this is suggesting to be placed under the hood. where is I2C or SPI there? One thing i'd suggest is other single wire protocols, like those used on digital temperature sensors

i have a couple of questions:
 - Are the inputs isolated between each other? the pcb photo on the website shows 8 optical fibers so it could be..
 many battery powered scopes are just floating, all the grounds are tied together. sometimes it's better to have all grounds separated, others to have them in common (lazyness) i don't know, maybe if it could be selectable....   
 - Ballpark price? We don't have any plan to retire our TPS2014 anytime soon, but i'd love to have longer captures, would make my life much easier sometimes

re: capacitive touch, i think it's only being cheap. only phone i had problems with when charging was a moto G
resistive touch is really a bummer unless it's multitouch (come on!)

[Yokogawa TM UK]

Q. Are the inputs isolated between each other? the pcb photo on the website shows 8 optical fibers so it could be..
 many battery powered scopes are just floating, all the grounds are tied together. sometimes it's better to have all grounds separated, others to have them in common (lazyness) i don't know, maybe if it could be selectable....
A. Every module is isolated channel to channel and channel ground too with the exception of two modules - one of the 16 channel voltage modules is only isolated from ground and the logic module is not isolated

Q. Ballpark price? We don't have any plan to retire our TPS2014 anytime soon, but i'd love to have longer captures, would make my life much easier sometimes
A. It would be interesting to get your opinion, some people who I have asked have said £3000 and some others have said £8000 for the chassis + modules. The TPS is very limited with its 2.5k memory per channel, we offer 100M per module which is then shared across the number of channels on that module.

Q. re: capacitive touch, i think it's only being cheap. only phone i had problems with when charging was a moto G
resistive touch is really a bummer unless it's multitouch (come on!)
A. Unlike the early resistive touch screens which made for a weaker comparison against the capacitive, you can pinch zoom on the DL350 touchscreen. The concept of a capacitive was explored yet was rejected as the measurement uncertainty would have been effected with noise in the most basic of test environments.

[JPortici]

so two point resistive touch. good

re:price.. i'd aim at the 3k mark too... 3k eur 

the selling points of the Tek TPS are battery powered and fully isolated inputs, then there is the tek tax (neglegible memory, very expensive)
when we have to probe sensors and busses inside a car we have a combo of
- laptop
- picoscope 2000
- kvaser canbus interface
which combined is between 2 and 3k (eur, not pounds!)

the downside of this setup, besides the spaghetti contruction, is that the picoscope has by design a significant blind time, because before triggering again it has to transfer the old acquisition through usb, also it doesn't trigger on serial decodes (because analysis is all made in the computer)

then for your product i wouldn't know how good the canbus module is, how many lanes can it scan simultaneously, if i can generate patterns to emulate modules, to test our modules etc

I really like the concept of this scope, it would be interesting to at least try one in the future (as i said, no plan to retire our scopes anytime soon)

[Yokogawa TM UK]

The CANBus module is captures the physical layer of the CAN signal, the user can then import a .dbc file for data alignment. You can measure 60 signals per port (2 ports), the module is fully isolated from ground and port-port isolation too and the sample rate of the module is 100ks/s.

Features such as wave window triggering and dual capture allow the user to capture transient events which could happen at any time over a long recording period. On other units you would either need a large memory bank or you would miss the event due to the timebase being too great.

Interesting opinion on the price

[nctnico]

Just for clarification: you can monitor 60 different CAN IDs? I suspect you are not talking about connecting 60 different CAN busses... 100ks/s sounds rather slow for a CAN bus so I guess you mean 100000 messages per second?

[JPortici]

need some clarification:
what do you mean by 100ks/s?

if by any chance you meant CANBUS running at a 100 kilobit per second speed.. that's almost unusable for today standards. basically it's only useful for the infotainment line.. and not for all manufacturers

60 different IDs per port sounds adequate

[sibeen]

OK, I'm envious, but then again I've been carting around a DL708 for about 17 years. Do you want to swap?

As to price, I'd guess at US$3K and then you tend to gouge a bit for the individual modules.

[nctnico]

OK, I'm envious, but then again I've been carting around a DL708 for about 17 years. Do you want to swap?

As to price, I'd guess at US$3K and then you tend to gouge a bit for the individual modules.

IMHO $3k should include at least one module otherwise it is like buying a car without wheels. In that case the car could cost $5 and the wheels $10k each.

Also this scope seems to be very specifically targeted at automotive so I don't quite get things like strain guage / acceleration sensor interfaces because for those I'd expect you will want way better FFT in order to look at precise resonance frequencies. And if you work on machines and sensors then I2C, SPI and UART decoding make a lot of sense as well because that would open up a lot more use cases for this oscilloscope (and thus increase the value for money).

[Yokogawa TM UK]

Hi All,

The CANBus monitors the physical layer of the signal and then a dbc file can then be imported post acquisition for packet data, the sampling of the physical layer is at 100ks/s.
It can be easy to confuse this with a simple oscilloscope - don't forget this is not just an oscilloscope - it is also a recording data acquisition system.
Call it a hybrid scope-corder...ah wait! 

The I2C, SPI and the like are part of our DLM2000 and 8 Channel DLM4000 oscilloscopes, if we had all the scope features on the DL350 then the argument on the existence of the DLM series would arise.

RRP is a little higher than the few opinions on here but then I suppose from this side of the fence it is expected

The chassis price is in the region of £5k (depending on where you buy it). Modules are priced separately because we do not assume to know what test you want to do, therefore allow you the flexbility of choosing your test.

It would be best to arrange a local demo so you can judge the unit for yourself - of course I am biased by saying it is brilliant but then I also echo the response from the last week alone which has been overwhelming!

[JPortici]

again, it's not clear what you mean with 100ksps. baud rate? messages per second?

The chassis price is in the region of £5k (depending on where you buy it). Modules are priced separately

pass. 5k including modules would have already been a stretch, but of course it's just my opinion i think R&S has something portable with the relevant options and specs in the 5k area...

[alm]

again, it's not clear what you mean with 100ksps. baud rate? messages per second?

The physical layer is sampling at 100 ksamples per second. Assuming it samples synchronous with the CAN bus, that would mean a max baud rate of 100 kbaud. If it is not synchronized, than even lower...

[Yokogawa TM UK]

I suppose it is down to what are you after, which is different for everyone - if it is a basic scope with 2 channels for voltage, current and serial bus then yeah go for it. The DLM2000 series starts at £3k so perhaps that would suit you better?

If though you are in an adapting test environment with changing applications then you may find yourself back in the market for a piece of test equipment that the basic scope can not offer.

The DL350 serves as the multi-functioning and modular solution that from an initial point of view may be seen as £5k + modules, yet customer who have already invested in the scopecorder series see the true value when looking further afield and seeing that buying several pieces of test equipment is much more than the all in one scopecorder.

I would encourage everyone to at least see the unit and then make your decision.

[nctnico]

Perhaps you should try and get the DLM2000 into the hands of someone who can do an in-depth review on it. BTW I used to own a DL1740 myself for a while and in my opinion it had a bunch of nifty features. I think a more general purpose oscilloscope is much more interesting than a very specific model catering to a niche market but it also depends on options prices. When buying a base model Lecroy or R&S (Hameg) you easely spend 50% to 100% extra on some basic options which makes the price way too high. In today's oscilloscope market you can buy cheaper models which pack a big punch. For example compare the R&S RTB2004 with the GW Instek MSO2000AE series. The first costs about 3500 euro more for (basically) a bigger screen but otherwise similar specs. People are going to wonder whether a bigger screen is worth 3500 euro extra.

[abraxa]

I think a more general purpose oscilloscope is much more interesting than a very specific model catering to a niche market but it also depends on options prices.

From what I gather from Yokogawa's portfolio (and its main clients), they appear to focus on those niches that other scope manufacturers didn't cater to before - assumingly because they know they don't have the same kind of resources that their general-purpose competitors have. Now that those, too, offer a lot of the features that were niche before (and thus were a unique selling point for Yokogawa), more niches have to be explored to avoid facing the competition head-on. Since most of us aren't part of these niches, the products aren't always attractive. It's also why I assume Yokogawa isn't too interested in the general-purpose segment.
Sure, the DLM series is certainly general-purpose if you don't get the niche options. However, those models are getting a bit long in the tooth by now - the DLM2000 came out in 2008 and has only seen a memory upgrade to 250 Msamples and more decoder options (CAN FD, PSI5, SENT). To be honest I was expecting a new scope in the DLM series instead of another scope corder but if that's one of the niches they want to serve then so be it.

Don't get me wrong, though. I love my scope and I consider it very well engineered, but as much as I wish it were different - if Yokogawa doesn't have the R&D resources to keep up with the other big players in the general-purpose scope market then they have no other choice but to choose their intended product audience wisely.

[nctnico]

I didn't know the DLM2000 series was that old. I wonder how that translate to the decode, math and FFT speeds. Recently I have been looking at a used DL9240 which is a 4 channel 1.5GHz scope but the price is still too high even after I put some serious effort into haggling. It has been discontinued for nearly a decade (according to the local Yokogawa office) so it if breaks it means money down the drain.

[abraxa]

I didn't know the DLM2000 series was that old. I wonder how that translate to the decode, math and FFT speeds.

I'm pretty sure they can't compete with recent models from the other big brands, given the age of the ScopeCORE ASIC and the core MCU. I don't think it was ever able to compete with the PC-based scopes but from my (limited) experience with other scopes it did compare favourably with integrated scopes of the day. The serial decoding was a big deciding factor for me and it does this very well. The Hameg and Tek models of the time were no competition at all.

Recently I have been looking at a used DL9240 which is a 4 channel 1.5GHz scope but the price is still too high even after I put some serious effort into haggling. It has been discontinued for nearly a decade (according to the local Yokogawa office) so it if breaks it means money down the drain.

The DL9000 line was started in 2005 and indeed it seems a little tricky to get a good deal on those even now. I would be very surprised if they don't have enough replacement parts in stock, though. I wouldn't be worried about them being unable to repair it.

[sibeen]

The chassis price is in the region of £5k (depending on where you buy it). Modules are priced separately

Been on a road trip, and actually saw some nice Yoko gear whilst I was on it.

I really think that your company is on drugs if you think that pricing is going to fly. Bloody hell, US$8 to 10K for a two channel handheld. Tell them they're dreaming.

[abraxa]

Been on a road trip, and actually saw some nice Yoko gear whilst I was on it.

I really think that your company is on drugs if you think that pricing is going to fly. Bloody hell, US$8 to 10K for a two channel handheld. Tell them they're dreaming.

While it may be too expensive for lab engineers like us, it's not necessarily too expensive for field engineers and service staff. Like I said, they know their clients and their needs, so if they determined that there's a need for a rugged, mobile chassis which allows the re-use the acqusition models for the scope corders the clients already own - it's a no-brainer to simply get a DL 350. It makes sense for those clients. Just not us. We have other needs