THD 82 GRAS EQ 40PM Miniature Production Line Microphone 1st Mic Using EQset technology.
foreign [Music] everybody Dave Lindberg in Hong Kong with another another episode of the THD podcast thanks for tuning in uh today we have a company from uh Copenhagen called gross and they do uh test microphones and they're owned by a larger company that does all kinds of things in test and measurement so we're going to be talking to them about a new product they have but first let's not forget our sponsor the Alti Association and they'll be hosting the Alti Expo in Orlando Florida on June 11th the 13th so we encourage anybody that happens to be close or willing to travel please go check that out uh it will be a great event to learn all about the different ulti members and their products and such so without delay let's get into the discussion here so Simon joining us from Japan how are you this afternoon Simon pretty good mate thanks a lot and Remy guastavino the director of product management from grass how are you this morning in Copenhagen I'm good thank you all right so yeah I think all of our viewers know the importance of having consistent and quality mics in their production test environment and uh I think Ross does mics for all kinds of testing but I think today we're gonna we're gonna talk about perhaps the the greater but maybe focused on this new product so um maybe introduce yourself and then we'll get into the discussion on the grass and the product that's my my name is Remy gustavino um I've been working for class four for two years now uh pre-order that I've been working for board on care which was also a microphone manufacturers for 12 years both in a research and development but also in product management and I've been working on small sensor okay thanks for joining us and maybe just to help spoon feed our customers we can jump directly into presentation and kind of walk them through the topic yeah let me do that yeah so we'll spoke about this new class AQ 40 p.m microphone that we've been working on for the past two years a bit about me but I've done it already so I could I could jump over this one so why is it so hard to perform acoustic measurement I've been working with acoustic for the past 20 years and I if I do to give a number which is that 70 to 80 percent of the measurement that people are performing are not valid and not do well so I will try to to help them measuring better that was my way of doing it and then I've been trying to focus on on the user Centric approach and there's two place to start you could start on on the user always is doing what we see choosing his microphone or you could look at the measurement chain the user is maybe something more for for University to teach on a company it's not we're not providing measurement lecture we can help them measuring but but it's not what we do so we work on the measurement chain if you look at a classical measurement chain you've got the microphone then you've got the node Geo analyzer and then you've got some kind of pcnn software and maybe some some cables but if you change your PC if you change your software if you look at measurement and certainly it doesn't mean almost anything because you get some small different if you're using different algorithm but usually it's it's pretty stable if you look at the sound card audio analyzer if you have something that is of good qualities and it doesn't mean any things are extremely stable extremely linear so that's not why we should work but if you look at microphone all microphone measurement microphone or production and microphone never sensitivity that is within plus minus 2 DB another frequency range that can differ up to 3 DB plus minus 3db so this is this is where we could make a different this is this is a place where you could optimize a measurement chain and then we had two that's two two kind of microphone production line microphone where cost is a key decision Factor where often the people using this microphone are not specially training acoustic so rely on an acoustic engineer and usually that's it it's not so often as they are playing correction such as text microphone edites sometime if if they are good enough oils that takes the time to do it if it's important for them they can set the sensitivity or read tests if they've got a dead sea enable uh user interface but they almost never correct for for frequency response only a few company do that and then you've got measurement microphone that I use in the lab where the quality is a key decision Factor where the user is usually highly trained and they are playing correction manually or with the help of a software so so we've that's why we decided to focus on production line microphone that's where we could make the biggest difference uh and typically you've got some some really short measurement chain and you're testing some simple device like a speaker a smart speaker a mobile phone some headset and like I said it's a key decision Factor when you're choosing a microphone on the production line is a cost so people have been using line microphone because they are an order of magnitude cheaper than a classical microphone measurement microphone and this is I've extracted 100 frequency response from database from production line microphone if you look at it if you look at the sensitivity so that's a that's something that would measure at 250 Hertz that's yeah you've got a spread that is within plus minus maybe 2 DB sorry I mean this is the microphone from the factory measured yeah exactly so that's what we get from the factory okay and and if you look at the frequency response you've got some microphones that have a big peak under high frequencies a microphone that are more flat some have a big belly there so if you don't apply correction you've got a measurement uncertainty that is huge especially in the high frequency range so that's that's really difficult and that's what we've been trying to address how can we help people on position like making better measurement like I said they need a better sensor but cost is really uh a decision Factor and and we've got some constraint we need it to keep the analog output because they want something that is Backward Compatible they don't want to change all the measurement setup they just want a better microphone we also need to to keep the form factor many many people on a production and are using this uh this kind of array microphone that are 34 millimeter long and seven millimeter diameter so they don't want something bigger or smaller they want the same size but it's a good thing for us is there's a lot of new technologies that are available that are affordable and that they're so small that we could put them into this microphone so if you look at it this microphone is unchanged on the outside it's got also an analog output it's got the same stick the same size as the same microphone but we were able to incorporate a small microprocessor inside the preamplifier so we've got a an analog input it's a condenser microphone and then inside the microphone we apply some kind of of digitalization then we can equalize so microphone to make it flat and then we can also set the gain so all microphone have the same sensitivity so which means that you've got a flat microphone with a constant gain out of the box and that's really important because it helps the customer credit cards cost because you could it's you've got a really fast setup if something breaks you can just exchange your microphone put a new one you reduce measurement and strategies your measure your reduced measurement complexity and you can use the exact same interface that you've been using before um usually if we come back to this 100 production line microphone you got them out of the box it looks like this and then you should individually adjust sensitivity so one way of doing it if you've got a text enable user interface analyzer but then it's really expensive and it takes time to to read all this chip inside the microphone to get the sensitivity and if you don't have a date enable analyzer then then you should manually type in sensitivity and then you always there's a risk of making a mistake but the worst boss is doing Equalization because you've got 100 different Equalization curve and you should have a software that support equalizing a microphone so that's really difficult this is a population of 100 of this new AQ 40 PM microphone using AQ set out of the box so no individually sensitivity adjustment no Equalization needed so if you look at your measurement uncertainties they are going much much smaller that does big plus hmm and that's really important especially on a production line because you limit the failed failure so if you got a good product that is for example underestimating high frequency that a good product could be classified as bad and and you should have to to trash it or to to revoke it but the worst is if you get a bad product but because you're not measuring a new right way it pass and then you're sending a bad unit to a customer and that's not something you wanted to yeah especially if it's a driver Factory and they respond they're sending that that driver on to final assembly and then it's it gets rejected at final assembly or even worse it goes into the assembly and then the final user might reject it and then it the costs amplify as it gets further down the supply chain I agree yeah so so it's really a a huge help to have a consistent measurement with small uncertainties yeah that's that's some kind of problem curve I took from this one on microphone if this if you're measuring this one as a big uh bump on the high frequency and the black one is going down so you've got more than two DBS difference between the microphone that's that's not something you want when you're measuring you want small incentives so Spain customer paying survey that you don't get different results using different microphone because your uncertainties are so small because the microphone is flat um this microphone is an order of magnitude more stable than classical production like microphone we've been working really hard on this new sensor element and and it's it's it's really stable which means that you don't have to compensate for for top yet or humidity or static pressure on a production line because of the stability of this cleansing element you don't have a long setup so you have much shorter setup time and replacement time and time is money on a production line so you want to be able to do sing quickly at your doesn't have an impact on your measurement uncertainties that's it's much better and then we were able to because we don't have to spend we spend a lot of time normally to tune the microphone response but this one we can tune it and then compensate for for all imperfection so it's a much faster uh Manufacturing I'm curious about one thing if you don't mind um uh the uh curve shape for the microphone yeah we pick up a few slides um is that the underlying characteristic of the microphone or is that a a characteristic of the way that it's measured this is a microphone okay and so then uh when you're equalizing the microphones you know in advance the ideal uh microphone characteristic the microphone shape that you uh equal into yeah yeah is it like a uh taking as a kind of a mean of production or is it developed in another way yeah so so we're we're working on on making it flat but but if you boost it or cut it too much and you you've got some repercussion on noise floor on THD so so so it's it's the mix-up of the thing we can do we want to equalize it so so it's it's extremely fat between 20 and 10 kilowatts yeah and then and then we cannot boost it too much because any will have an impact on thdn and we don't want that sure and um then uh there's there's a little bump around 10K right at the top of the range and um is there uh you don't want to equalize that to perfectly flat versus is there a constraint around doing that oh it it's a it's a small bun but but if you look at the scale it's almost extremely small scale so this this microphone is within plus minus 0.5 DB from 20 years to 10 kilowatts well I see okay yeah it's plus 2 minus 3 usually when you see a frequency response is the plus 10 minus 10 or so it's an extremely zoomed range I I use Now understand that understand that um I just curious partly from the point of view of saying that uh if uh the users have already got a system without expecting this uh uh exact microphone characteristic you wouldn't actually attempt to uh equalize say around the 10K Point down by a point something of a DB because that would make it a different microphone is it yeah we can we can do that but again it's a it's a production line microphone so it optimized on on cost like the equivalent if you take a grass 40pk or burning care 4958 so which was the older generation of measurement microphones that are plus minus three DB and a plus minus two between 50 and and 10K um so so we we've been crushing this this and then this from plus minus two let's say if I'm plus minus two plus minus 0.5 yeah and then if you need something better after 10 kilohertz then maybe you should use a measurement microphone because this microphone is a an order of menu to cheaper than a measurement microphone so so you you cannot you cannot get it online you can't have everything yeah okay and then so another part of the design that you mentioned is um is uh trying to make the microphone stable um and uh that includes meaning stable over a long period of time it does we've been we've got some some things that we've been working on for more than two years and it's it's uh they didn't change since sensitivity or frequency response if you're looking at older generation production line microphone only if you look at that the humidity they could move more than plus minus one One DB insensitivity because it's your middle because it's warm and then because I've got this is still races so they don't come back to the same level yeah and it takes time so you've got this time constant it's it's a it's it's it's very difficult this microphone is if you ask me it's it's a much much better microphone and do you think a lot of your customers uh do a regular uh calibration check or many actually don't bother I think it's depending if you're talking about high-end uh audio devices then maybe they're looking at doing more calibration but I can also see that the quality of audio devices of headphones and headset and mobile phone is rapidly increasing and and the price is also not moving so much it's increasing but not yet so I think that won't really need a better production microphone because you've got some match I mean if you're buying a really expensive headset that goes to 700 then you expect it to sound good and then they've got some ANC microphone that they should match and then you need to have a reference microphone that is stable and a good sensitivity so you could adjust your microphone so I really think this is needed and I should see the first reaction from our customer is is exactly happy about it and then everybody has kids they could test it and have you been sending some some so many samples that we don't we have to polish more now okay that's good yeah that's really good yeah so so what did the island or what did we learn we learned that um the measurement one on political name is is very conservative so if something is is working you shouldn't change it and that's why we had to do this constraint about save and about uh plug and and way of pulling your microphone we can also see that in 2023 most used acoustic sensors are still analog nobody is using digital sensor yeah at the beginning I was really preaching for for kind of USB sensor or a Wi-Fi sensor but people are not ready to do that yet and that's why we decided to to incorporate all this technology and and inside the problem with fire but still you'll see analog output but then we also show that we could make a real difference to for the user by buying component this new digital technology in an analog sensor and and it's the first step on uh on the acoustic sensor technology measurement microphone I've been pretty much like the development of the measurement microphone has been so exciting for like 20 years let's say so so this one is the first step and then we've got a lot of new ideas that we will apply to to microphone and like if you should remember something that we've been working on this microphone too is the the use the user to to reduce measurement uncertainties so that people can focus on their measurement of working on the singer doing and and forget about microphone techniques um okay good stuff one more one more quick question we've talked to uh it's mostly what you've talked about is focused on uh the uh consistency of the microphone um can you comment a little bit on some of the other characteristics dynamic range uh noise floor and things like that so some uh some quick fact about it it's a it's a 20 kilowatts uh pressure field microphone it's extremely price competitive compared to to a measurement microphone it's a CCP or ICP or ccld per word so so it's worked for for any microphone sound card but it's also designed so you can use it directly on a 48 volt phantom power sound gun and that's really interesting because you don't have to set sensitivity because it is a like and you don't have to correct for frequency range so you could have a a measurement chain constituting of 200 audio song counter cable and then your microphone so you could have a measurement change with extremely low uncertainties for the fraction of a price you will pay for classical measurement change so it's specif it's 34 millimeter long it's seven millimeter wide it's a 20 at 20 kilohertz the pressure response is within plus minus 0.5 DB 20 hours to 10 kilohertz and then 10 hours to 20 kilohertz plus minus 2 DB the sensitivity is 25 millivolt per Pascal plus minus 0.2 DB so that's also usually it's plus minus maybe 2 or 3D
so we go all the way to 0.2 because we can correct for getting so noise flow is lower than 30 DBA the acoustic overload point is higher than 120 DBS Bell with less than three percent DHT and the environmental stability is better than 0.3 DB and what I mean by that is if on a on a normal production line you can assume that the temperature will be between 15 degrees Celsius and maybe 40 degrees Celsius and the the parametric pressure won't change more than plus minus 30 ectopascal and it won't be raining inside so if you put all this parameter together the sensibility won't change more than 0.2 DB that's what we but that's where we are okay fantastic it's a it's a really nice and I'm so proud of this microphone we we managed to achieve extremely good specs can I ask also just uh um the uh the front housing uh and the previous microphone you showed it has the metal Grille with a lot of slots I don't call that this one slightly different designers there's something to that is that a yeah it's because you're using this microphone really often to to measure um maybe an iPad speaker or the music to measure a men's microphone on a on a mobile phone and then if you're measuring a small a small microphone or a small speaker you don't want to measure it with a with a huge uh sensor so so as a world of rule of thumb you should you always use a sensor that is in the same Dimension or smaller than the thing you're measuring so so we we have this new uh optimize uh sound ports here that that is really well suited to measuring small stuff very close by okay so you kind of hinted that you've been sampling that already is it and uh so is it in mass production yet or is it that coming soon what's the timeline for people to start ordering this for their factories yeah so so we are mass polishing it now okay uh we had a pre-production series but all the same bowler have been sent to to companies that want to evaluate it but we've we I think we will be delivering in in a week we will have some a new batch coming and then it will be much much prediction okay all right very good and uh Simon did you have any more questions I think I got everything I wanted here cut everything in already all right um yeah let me think uh so yeah ready to ship um yeah I'm not sure is there anything else that you'd like to add Remy before we uh wrap this up no I think we've got a lot of good information and uh yeah the plastic those messages that we we made a price competitive microphone is that can can we set like we don't if you can tell numbers great if not can you talk like you know percentage less than uh than like the your predecessor microphone of similar yeah so so it's on the same price range at the classical uh production 9 microphone so the price is almost unchanged and it will be a register around the third of a price of a measurement microphone right with almost the same kind of specs right the the dynamic range is a bit smaller but you can measure anywhere up to 120 so it's not it's enough for Consumer products exactly okay so if if you don't have to go super high frequency or super low noise floor or super highest build it's a perfect microphone okay all right well we'll put some links to the product in the description below and we encourage people to check that out and uh of course like everybody like subscribe uh whether you're Spotify YouTube rumble or apple podcast wherever you're listening to this or watching it please give us a like we appreciate that and Remy thank you so much for joining us today and uh thank you and always I'm always with his uh curveball questions there that's always fun hope you enjoyed that and uh so yeah thanks again and we'll see you next time everybody thank you
2023-04-03 01:47
