This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TLV320AIC34EVM-K: Purepath configurator

Part Number: TLV320AIC34EVM-K
Other Parts Discussed in Thread: PUREPATHSTUDIO, TLV320AIC34, CC8530, PUREPATHCONSOLE, INA1650, DRV135, OPA2320, OPA320, TLV320AIC32, CC8520, TLV320AIC3204, CC2592, CC2590

I have tried using this so called tool for hours and making no sense out of it. The examples are not what I need to do. It appears that you can't dynamically change the device of choice in the tool. Well, you can but it changes nothing. I have searched for some video tutorial on using purepath but can't find any. I just want a 4 line input to 4 line output end to end. Is it possible that I can write my own code to configure the devices with code composer studio? and then use cc debugger to program it? Of all the TI tools I have ever tried to use, this one is the worst. The help file doesn't help. It just shows what you already see. Very little in the tool is intuitive. I am a 40 years electronics engineer. I designed with TI BLE -4 devices in the past. I already have the hardware designed, just need the software. I see no useful notes on the TI site concerning this. If you don't want me to use TI parts, this is a good way to help me choose. I should have started my design with the tool to see how bad it is. Anybody give me some help?

  • Hi Paul, 

    I'm sorry you are having some problems.  Can you clarify what tool you are using?  the title of your thread mentions the TLV320AIC34EVM-K  but that EVM doesn't use Purepath studio or purepath console. 

    best regards,

    -Steve Wilson

  • Hi Paul,

    Agreed with Steve's comments that it's never good for us to hear of customers having challenges like this. Could you clarify what you purchased and what tools you downloaded?  PurePathStudio (PPS) is what it sounds like you're using, but the TLV320AIC34 does not have a mini-DSP inside it and therefore does not interact with PPS. 

  • Hello Paul, 

    I would like to reach out again, and just say that we are committed to helping you. We do need some more information regarding which tool and device you are using.  

    "Purepath" is a TI trademark that is often used with the Audio products and associated software.  This unfortunately can cause some confusion when working with different devices and software packages. There are 4 software packages that use the Purepath name:

    Purepath Studio - (PPS) Used for programming the MiniDSP in our portable Audio Codecs, ADCs, DACs and Amplifiers. 

    Purepath Console - (PPC) Used for smart amp tuning for amplifiers, and more recently for configuring the TLV320ADCx140 ADCs

    Purepath Wireless Configurator - (PPW-CFG) used for configuring the CC85xx series wireless devices. 

    Purepath Wireless Commander - (PPW-CMD) Also used for the CC85xx series devices

    Personally I am most familiar with the top two (PPS,PPC),  but I have used the wireless software as well.  none of these software packages will control the TLV320AIC34EVM-K which you mentioned in the title of your thread,  which is why we need a little more information. 

    While I would like to say that these software packages are flawless and simple to use even for a new engineer, the reality is that there is a learning curve for some of them and none of them are perfect.  That being said we are here to help.  I check the forum multiple times a day and would be happy to help you in any way that I can.  I am responsible for the TLV320AIC34, and support devices that use PPC and PPS.  

    If you need help with the wireless products, I can help connect you with the support team for those.  

    best regards,

    -Steve Wilson

  • I was trying to use the PP tool and yes it does support the TLV320AIC34 as that is the audio CODEC I selected in the tool itself. I am using the cc8530 with it;
    I am not an idiot.
    I haven't purchased anything until I can formulate all aspects of the potential design. I have designed well over 10 products in my experience for Caterpillar and other companies.
    I worked with TI products on many occasions. Back in 2014 I designed a BLE keyfob using a TI device.
    I need to see that the system components can play together, parts are available, what the component costs are, that a PCB can be laid out in a reasonable size and layer count.
    Obvious as well that it can perform under all the conditions needed. Once a clear design pathway can be put together in concept, can items be purchased. Right now I am not convinced.
    From the PP screen;
    [TEXAS INSTRUMENTS TLV320AIC34 - low power 4-channel audio codec

    DEFAULT SETUP (I2C):

    Serial format: I2S, 16/24-bit, continuous clock, no resync, MCLK generated by CC85xx (AIC34 PLL is not used).

    A/B inputs: Differential line in (L1), configuration of input volume control "fixed offset" should be set to +6 dB.]
    I interpret that the code generated by the PP tool runs on the CC8530 and communicates to the 4 channel codec through the I2C interface. So I don't see any code needed internal for the CODEC itself.
    But it does appear to need to be configured. It furthermore appears that the CC8530 is flash programmed through the SPI interface. Whether code debugging can also take place through SPI is another subject.
    It appears that if we agree on terminology, that a so called "Master" unit consisting of the CC8530 and AIC34 (Along with supporting passive components), would in my case be set up to broadcast 4 audio channels. 
    In my case, the slave would also consist of a second set of the CC8530 and AIC34 (Along with supporting passives), to then output the same 4 audio channels, wirelessly.
    The issue is that this setup is not obvious to me in the PP tool. It seems that the slave is only intended to be one or two, of the four channels.
    The other uncertainty on my part is the use of the Volume up down, forward reverse IO pins on the CC8530. I assume the names on the Master and Slave are the same. That the selection of IO pins can be assigned to any allowable IO pins. That if the master IO pin selected pulses low, that the same assigned pin on the slave will also pulse low after some time delay.
    Can you please make comment as to how I might configure the CC8530 and AIC34 using PP for my desired 4 channel in/out use?
  • The tool tile is Purepath Wireless Configurator 1.4.2.38775

    Can't I attach images in this forum tool?

  • Hi Paul,

    Thank you for the additional information.  We thought you were asking about the Audio software our particular Audio team supports, which is PurePathConsole, and PurePathStudio.  Steve is correct that the TLV320AIC34 is not included in those tools. 

    We're admittedly a little less familiar with the wireless configurator but will reach out to the CC8530 team to pull them in for some help as well on those questions as well.

    Yes, you can attach images to the E2E forum.  Select the "Insert/Edit Media" button in the editor and then select your file.  Unfortunately dragging/dropping or copy/pasting images directly into the editor is not currently supported.

  • Before I do attach any schematics or images, is this posting public or private?

    I don't mean to sound hostile, I was agitated when I first inquired about the TLV320AIC34EVM-K: Purepath configurator, as I struggled with it for a couple hours, and couldn't get it to do I think it should do.

    I was unaware that TI had multiple software programs / user interfaces with the same name. I can hardly be blamed for that.

    Since the CC85xx devices are wireless audio transceivers, it only stands to reason it is an audio product. Yet it is also wireless, so it is also that.

  • Hi Paul,

    I'm sorry for all the confusion about the naming, I to sometime get confused by the "PurePath Wireless" and the other audio ICs that also feature something called "PurePath".

    Officially we only support the application roles listed in the PurePath Wireless Configuration tool. This means we only support a 2 ch slave output. It could potentially be possible for you to add in your own application role to enable four channels, I can verify that you could set up such a project to some degree. 

    Unfortunately, I do not have any hardware at the moment which I could use to test this out, all I have is stereo codec setups which is not much help when testing the 4ch hack. What I can do is to provide you with my custom application role file that you could try out yourself assuming you got the hardware to test it out. 

    The file attached to this thread should be placed at the installation folder inside "application_roles". After a restart, you should be able to select 4ch output on the slave device. However, as I mentioned above, I can't tell you if or how well this would work as I'm not able to try this out any time soon.

     /cfs-file/__key/communityserver-discussions-components-files/6/s_5F00_4ch_5F00_output_5F00_satellite.ppward

  • Thank you for your response. And for the file. I will give it a go. 

    I seen that Qualcomm has a stereo Bluetooth device that has an integrated Audio CODEC. Do you know of any new such devices that TI is working on?

    I tried to attach two files to this message. But it doesn't look like it attached correct. One is a 3d image of my PCB layout, the other, the schematic. So I don't know if you have a way to retrieve them.

    I have a couple questions about the pinout on the 4 channel codec; I used Line 1 Left & Right, Minus and Plus for all four channels going in on the Master through DRV135's. Then I used Left & Right LO pins for the Slave side output audio through the INA1650's. Does that seem correct, or OK?

    What should I do with all of the unused microphone inputs, Din, mono, and Line2 inputs? Can they be left floating, or tie them to Gnd?

    Thanks

    BLUE4ch.pdf

  • Hi Paul,

    I'm not aware of any such devices. There is a few examples on how stream audio (voice in this case) over BLE but these devices (cc26xx) is also without codec and only got I2S outputs.

    I'm sorry but I'm not very familiar with this codec or the audio ICs in question so I actually don't have any good answer to give you (that would not risk me giving you bad advice). Hopefully some of my colleges from before can pick up and help you on these questions as they do not directly relate to the CC853x part. I'm mainly focused on the wireless technologies and do not want to risk give you any bad advice on how to work the codec.

    Looking at you schematic however, I can see that you have two CC8530 mounted on the board (as well as two codecs). Could you maybe give me a run down on how these are to be used in the actual product? Reason I ask is because there is some considerations related to operating two CC853x on the same board (I could not see the model you attached so I don't know how your board actually looks).

    The consideration is mainly focused on timeslot alignment where you can setup a physical signal between the two devices to keep the TX and RX in sync. This to assure that you do not end up having two masters close to each other drifting out of sync as this would mean you would get to the point where one of the masters is shouting into the ear of the other, degrading performance. 

  • Hello,

    Here are a few notes on the schematic:

    1.) The DRV connections look appropriate, but beware that if they output a +/-7V output based on their supply voltages that they'll break the TLV320AIC34 by over-voltaging the inputs.  When required, many customers use lower voltage single-ended discrete receivers that are powered from the same analog supply as the AIC34 to avoid over-voltage condition.  The connections to the CODEC crossed around but it looked like they were all connected properly.

    2.)  The INA1650 connections also look appropriate.

    3.)  You are shorting DOUT to DIN for all of the CODECs.  You will need separate connections for DOUT and DIN as shown in the application section drawings of the datasheet if you intend to control the DAC outputs from a source other than the DOUT signal coming out of the ADCs.  

  • 1) I am glad you noticed the voltage max issue. I figured that since the DRV135's need min +-4.5 volts, that 5 to 7 volts would work for those. I figured the Line input wouldn't ever exceed normal line levels. I looked at discrete op amp designs but they need matched resistors or error offsets get introduced. What about a series resistor and a 3 volt zener on the inputs of the AIC34? 

    3) I have 0 ohm resistors that appear to short the in to out on the A and B sections of each AIC34. The same PCB would be used as a Master Transmit or a slave receiver. By populating one or the other 0 ohm, I was expecting the data direction to be receive or transmit from the perspective of the CC8530. I was unable to reason any other configuration. I figured that if the software configures the CC8530 as a transmitter, that the data is an input on the CC8530. And vice versa for if the CC8530 is a receiver. If this thinking is wrong please enlighten me. There seems to be 3 ADx pins on the CC8530. 0, 1, 2. I see no examples of 2 being used. I don't see a good explanation of this interface in any of the datasheets that I have.

    You mentioned something about having two CC8530's on the same PCB. I was planning to share the one 48MHz crystal between the two. Is there an issue I need to be concerned about?

    The PCB is about 4.8 inches by under 2 inches wide. This board is 4 layers. My first time laying out such a board. Normally companies I worked for had a team of people that laid out the PCBs. I used a free open source tool called KiCAD. The image is from the 3D viewer. That comes in real handy. I don't figure on populating those headers. Perhaps soldering wires in the holes. and using pogo pins to program.

    I might have to have someone solder the BGA's and CC8530's on at least, if I go this route.

  • Hello,

    Thanks for the notes.

    1.) Agreed there shouldn't normally be an overvoltage condition, but should the inputs to the DRV get overdriven, or something strange happen on startup/powerdown it could output a voltage large enough to damage the ADC.  Your solution with a series resistor and some clamps is effective.

    2.)  This information makes sense from the CODEC standpoint, but will need a second set of eyes from the CC folks.

  • I found an example circuit in a TI website named SBAA265–October 2018

    It shows a single end to differential circuit using an OPA320. I found a dual OPA2320 for half the price of the DRV135's I will need .1% resistors, but still cheaper.

    The OPA2320 has rail to rail output and works down to 1.8V single ended. That leads to a question; I plan to power it from 3.3V. But there is a reference voltage on the non-inverting input of the inverting side of the differential amp to the M input on the A2D. The example circiuit is powered from 5 volts. so the reference is half of that. I am thinking that if my circuit is 3.3 volts that I could use the 1.8 volts that also powers the DVDD pin on the AIC34. Digital core voltage supply, 1.65 V to 1.95 V. Is there any connection in the core voltage (DVDD) being about half the 3.3 volts? It appears that the operating currents on DVDD are minuscule. 

    Table 5 in SLAS538b shows; 

    CM SETTING     RECOMMENDED AVDD,DRVDD      RECOMMENDED DVDD
    1.35 V                             2.7 V to 3.6 V                                1.65 V to 1.95 V
    1.5 V                               3 V to 3.6 V                                   1.65 V to 1.95 V
    1.65 V                             3.3 V to 3.6 V                                1.8 V to 1.95 V
    1.8 V                                3.6 V                                            1.95 V

    It doesn't appear that this is an output available for me to use. However does the PP software tool have some control for that as it says in the text? I assume it would be an I2C command sequence.

    Should I provide a resistor divider on the 3.3V or is 1.8V close enough? Half of 3.3 is 1.65.

    Also don't forget your comment on having two CC8530's close to each other. I have a few little China made Bluetooth transceivers I was playing with, I set up 3 sets, 3 pairings of 6 units. and sent audio through them. They seemed to synch to each other and I had different music going through them.Seemed OK. They are not on the same PCB.

  • OK another frustration, I just typed a whole bunch of stuff and it disappeared. Such is life.

    The tool update code you sent does let me create a 4 channel satellite. But no GPIO options for rewind FF vol up and down.. GPIO12 through 15 are not available even without the range extender option.

    Where are the settings in the PP tool for Line 1 inputs verses line 2 etc?

    Once you select certain things in the PP tool you can't change them, you have to start over. At least as far as I can see.

    I appreciate you sending me the 4 channel file, but how can we make it official?

    If I spend time and money on this project going forward I need a little more confidence that the tool will support it.

    I don't blame you, but if TI sells a part, the user needs support for it.

  • Hi Paul,

    Text magically disappearing is one of the frustrations of life, I can relate to this.

    As for GPIO options, you would need to select which "Remote Control" commands to enable for them to show up in the IO mapping. This is done in the "User Interface" view, there should be a small checkbox in the list of Remote Control selection.

    If all GPIOs is not available in the IO Mapping view that means that your configuration is setup to reference one of the evaluation platforms as HW. If you go to the configuration view (same as where you select application role) you will find a "Hardware Platform" drop down in the bottom. Changing this to "Custom design" enables all IO Mappings options.

    As for the "select line 1 vs line 2 in the tool", could you elaborate on what you are going for exactly? If you refer to AD0/1, this is selected in the Audio Streaming view where you bind logical channels to audio IO.

    On the note of the 4ch mod I shared with you, it is not likely that it will be made "official" (I'm sorry). What is in the tool today is what officially is supported and it is designed to cover the most common use-cases. Unless there is a big market opportunity to but the resources into the full verification to make it "official", it will likely never happen. 

    That said, support is not directly limited to this pre-defined set of options in the tool. It simply gives you a set of things that TI says "will work good out of the box". The option to tweak the settings to your own need is still possible (the 4ch role being an example) but it comes at the cost of you having to verify the performance yourself and TI not guaranteeing any tested numbers. We (I) can support you here on E2E on how to do so but I can't really impact any decisions for future development of the PP tool as such.

    On the design note of having the two devices close to each other on the same board, this is fine but in that case you should really use the master/slave timing signal. Consider it a coexistence improvement, you can do without it but you will likely have performance issues as the two radios drift apart so that TX and RX align. As for driving both with the same crystal, I would not recommend this as the device in question is driving the crystal. 

  • I appreciate your personal assistance with this. So I am bold enough to guess that I have no competition in my design/product idea? If no one else is inquiring to do what I am?

    You; As for the "select line 1 vs line 2 in the tool", could you elaborate on what you are going for exactly? Yes,

    Me; My assumption is that the CC8530 in my case, as an I2C master, writes commands to the slave AIC34 probably just after reset. The AIC34 has many options as Microphone inputs, bias, Line1 (pins; J10, J11, J10, J11, K10, K11, K8, K9),  Line2 (pins; E10, E11, F10, F11, G10, G11, H8, H9), There has to be some control as to which of those does what. (9.3.18 Audio Analog Inputs) I assume it is through the I2C, as programmed by the PP tool in the CC8530. So the PP tool I would expect to let the user define the setup in the CODEC. If I am wrong, then how do those pins get defined? I am using Line 1, so I need to turn those on and Line 2 off. I am not using headphone or other stuff, so I need to turn those off. There are quite a few register settings in the CODEC for all kinds of stuff. (Table 5. Analog Output Common-Mode Recommended Settings) - how to set that, all from SLAS538B –OCTOBER 2007–REVISED NOVEMBER 2016. There must be 127 registers to be set in the CODEC.

    From the SLAS538B document; 9.5 Programming
    9.5.1 Digital Control Serial Interface
    The TLV320AIC34 is entirely controlled by registers, with a register map that is software compatible with the low power stereo audio codecs TLV320AIC3x and TLV320AIC310x. To maintain best software compatibility with
    stereo codecs, the register configuration of the four-channel TLV320AIC34 is divided into two separate I2C slave
    devices containing separate addresses, with each address used to access registers controlling two channels of
    codec and associated inputs and outputs. The two partitions of the device are denoted A and B, with analog and
    digital inputs, outputs, and internal blocks named accordingly, ending in _A or _B. The two I2C addresses are
    also denoted A and B, with each used to control the correspondingly named signals and internal blocks.

    Me; OK where are these A & B I2C buses? I only see one. ADDR_A and ADDR_B must then be used to determine which A or B side is being written to. How does the CC8530 control that?

    Otherwise some other processor would need to write all the set up commands to the TLV320AIC34's. Is that where this is heading?

    You; AD0/1, this is selected in the Audio Streaming view   

    Me; Yes I see that one, and have played with it.

    You; On the design note of having the two devices close to each other on the same board...

    Me; I am still a little confused. But just to make sure we are on the same page, looking at one of these 8 channel circuit boards as I sent you a picture of, the whole board is either transmit 8 audio channels, or receive 8 audio channels. The transmit side sends also the rewind, FF, volume signals.

    The cc8530 datasheet does show the crystal pin as capable of taking in a clock on pin 21. Crystal oscillator pin input, or external clock input (48 MHz).

    The other question is overall system delay. If the delay through the entire end to end audio is more than ~30mS, this could be a problem.  

     

  • Hi Paul,

    There is a few things to consider, as for your idea being unique I can honestly not say. Just picking something from thin air I would say the use-case is typically more of a remote speaker/microphone one. This means you typically have one (or two for 8 channels) masters sending to multiple slaves where each slave is their own physical speaker. The "sound bridge" application is valid, i just don't think it is the most mainstream use of these devices :)

    OK, now it is clear that your question was regarding the configuration of the device. This is done from the PP tool as well and you are free to alter the setup sequence as you see fit. To do this, you first need to set "Custom Setup" to "Enabled" in the Audio Interface view. This enables another view, "Audio Device Customization". In this view, if you start by clicking the "Set all sequence to default" it will populate the view with the default configuration used to setup the device. 

    You can alter these views as needed. The configuration is divided up in several section/transitions which relates to the device state. For example going from "Off" to "Sample Rate switch" of from "Active" to "Low-Power". There transitions is in relation to the PP device and not the CODEC. For example, "OFF to SR-Switch" is the PP device booting up and covers most of the basic setup of the CODEC.

    As for the board layout question, my feedback is related to the protocol. No matter if the board is either TX or RX in the sense of audio direction, the radio protocol will still be bi-directional. This is why the sync is important to assure that the masters do not drift apart (this only applies to the master side) as the optimum is them performing TX and RX in sync. 

    You could most likely drive two CC8530 using an external clock source, you specifically mentioned using a crystal which would (likely) not work as well. With external clock a buffered oscillator is typically assumed, something that is capable of driver one or multiple outputs. Even in the case where would assume we are OK with setting one of the devices up as the "driver" and the other only takes the input of the crystal, the extra load on the crystal could lead to issues. 

    As for audio delay, this is dictated by the "Audio Latency" under the masters "Audio Streaming" view. the default is typically 2048 samples (maximum) and the resulting latency in ms is given next to it (it varies with sample rate used). You could lower the latency by reducing samples, note that doing this affects your network robustness. Basically, smaller sample buffers gives you lower latency but at the cost of not having this extra headroom to recover from possible radio issues (dropped packets etc). 

    In the end, the final network performance depend on several factors, most of them boil down to the time that is left to handle communication errors and recover:

    * Numbers of slaves in the network, in this case i assume you only have one and you could not really go lower.

    * Numbers of channels / data to be sent. Logically, sending less data means having more time for correction

    * Sample buffers size. Larger buffers means you have a bigger window for recovery without the audio dropping out.

    * Data rate and protocol (these should really be the default in your case, 2 Mbps and "B" protocol or more of a mono/microphone feature. 

    * Audio "fade-out" before muting. This buys you a few extra buffer slots if it is turned off but at the cost of a possible "pop" int he speaker if you actually have an audio drop. 

    * Multiple master timeslot alignment. Basically what I have been talking about before, making sure that the masters to no "shout" into each others by aligning TX and RX operations. 

  • Strange thing. The TLV320AIC34 no longer appears on the TI website. The TLV320AIC32 does.

    What does this mean? Just looking for the data sheet and it isn't to be found. Even the Digikey website link doesn't work. Digikey still lists the part.

    https://www.ti.com/audio-ic/converters/codec/products.html#p92=1;8&p1340=I2S

  • Hello,

    I think this is an error in the product search.  The device is still active and is our best option for a 4-channel IN/OUT CODEC so it's certainly not going anywhere.  

    Here is a direct link to the product folder.  

    https://www.ti.com/product/TLV320AIC34

  • I was expecting some oddity with the TI website. But it just seemed strange. Thanks for the link.

    So I did find the I2C deal in the PP tool. Looks like I can set behaviors for CODEC, turn On/Off, and so forth. To elaborate on my observation in this I2C deal, it appears that the "Default" commands that pop up in the tool seem to be set for a two channel CODEC. So I am guessing that I will have some coding to do for use with the 4 channel one. I have experience with I2C, and I understand coding. So it looks like it shouldn't be too difficult to figure out. I may have to go through a few iterations once I have hardware to play with.

    Since you modified and sent me that script, is there any other anomalies that you can think of with this? Will a proper .hex file be created by the tool for the 4 channel CODEC?

    Is there any special reason that both Master and Slave configurations be in one project? Is it just easier to look back and forth at them. I made two projects, one for Master, one for Slave.

    I did find a small oscillator 50 ppm, and have that now in the schematic. I placed series 33 ohm resistors, one for each destination (cc_8530), then matched the trace lengths. I also made sure the GND returns have no gaps on that run. I don't know the output impedance of the oscillator right now, for the value of the resistors.

    It appears that some of the differential lines can be used single ended. Is this a whole lot less quality audio signal? Is it worth me using the differential to single external components?

    Thanks

  • Hi Paul,

    It is hard to say, I myself have not been able to try this out as I lack the hardware to do so which mean I can't tell to what degree it runs. I'm not sure the timings have been checked for the 4ch slave (the slave need to prepare responses to the master). I would not be surprised if it work but with the network performance (read range) taking a potential hit (as one would expect when splitting the available BW over more nodes). I got the feeling that you already have the AIC34 dev kit, as that so? If that is the case, do you also have a CC8530 kit?

    Having both Master and Slave in the same project is a convenience thing really.

    I would like to further stretch adding in the timing master line on the boards to assure that they remain in sync.

    As for the differential line question, I assume this was directed towards the CODEC?

  • Hello,

    The Line-Out pins on the CODEC are really only meant for differential output operation.  If you'd like a dedicated single-ended output, consider using the HPOUT pins instead.

    We had a related question a day or two ago where our AIC34 lead replied:

    https://e2e.ti.com/support/audio/f/6/t/908164

  • I have zero TI, BT hardware. I see that there are eval boards for the 4 channel CODEC and the 4 channel Transceiver. My first stab at this 4/8 channel line to line idea is to conceptualize based on data sheets, and seeing what compiler for software is needed. I have enough experience designing products to conceptualize to see if there is an appearance of feasibility. I wouldn't spend money if I couldn't get through this first phase. Then If it appears doable, then I might consider eval boards. But I would want Master and slave sides, so that doubles the costs. I would guess that to be around $600. Then if I did my own layout and a well thought through plan, with conversation with you (the expert), I could spend about the same for actual hardware for my own purpose. I have PC's, Freq counter, scope, DMM, bench supplies etc. At that point I might just go with my own board layout. I am retired at this point, but not dead. I am hoping this doesn't effect you helping me. If this pans out, I can sell this product. Perhaps not the volumes of a Caterpillar, but sales non the less. Who knows, a bigger company could buy my idea.

    What are the timing master lines?  Here is all I found; The fifth FW release (FW 1.4) adds support for protocol master timeslot synchronization, frequency planning and proprietary remote control/custom-defined USB HID reports. This enables, among other, the following new applications: Multi-protocol master systems, e.g. base stations containing multiple CC8530s to serve more protocol, slaves and/or audio channels (CC8530/CC8520) where is this explained?

    I also am assuming that the cc8530 will completely be able to configure the AIC34 through the I2C interface. I hope that the totality of I2C commands for 4 channels will not over fill the ROM in the cc8530.

    As for working in the presence of other BT devices, I assume the following bullet is true; Excellent robustness and co-existence through multiple techniques

    As for the differential line question, I assume this was directed towards the CODEC? Yes, I can stay differential, just curious.

  • Hi Paul,

    E2E does not go anywhere, we still do our best to support no matter if it is custom designed hardware or evaluation kits. The reason I asked was simply to see if you had the possibility to try it out as I myself lack this until I can again return to office (and that is unclear with the current situation).

    As for the timeslot alignment, you can find a good summery of this in the help section of the PP tool (F1 (to open help) -> Project Panel -> Advanced Option Panel). IN practice, it means setting one up as a "timing master" and the other(s) as a "timing slave". This will give you an additional IO in the mapping table, "Timeslot alignment control output/input". You then simply connect output of the master with the input of the slave(s).

    I would need to check any actual numbers for the I2C command list but i would not assume this to be an issue as it would most likely be some headroom in flash for quite big I2C command list.

    Co-existence is always a tricky question to answer and the short answer is yes, the device should be able co-exist with other networks of different technologies but it all depends on the exact nature you operate the devices in.

    For example blasting a WiFi router right next to a BLE device could cause issues even if they do the best to avoid each other which leads to a whole other topic of protocol co-existence :)

    I assume what you are looking for are more of a "there is other networks in the air" and not "I'm stacking this on top of my High Power 2.4 GHz WifI router"? In that case, you can find information on what the protocol does in terms of frequency hopping etc in the CC85xx Family User's Guide. 

  • I know that you don't know my background, and I don't know yours. So anything I write to you I don't intend to sound like a know-it-all, because I don't. But I am yet to not understand anything you have said. The only issue being details on the devices, and interaction specifications unique to them, and where to find them. For this info is what I seek. I am familiar with EMC testing and RF to some extent. Though at times trying to remove some RF emission from a product, can seem like black magic. I have spent days in/at Anechoic chambers, field testing sites. I had to do emissions, susceptibility testing, and Electrostatic testing. Power-cross testing, Power surge/burst, UL etc. That at my years at Rockwell, and Westell.  Cat had their own EMC facility in Peoria, IL. My years at Caterpillar were great from the standpoint of TI experts offering half day, or all day classes on various TI products and tools (Other vendors as well). We had a day on the TI MSP430. I have also had a 3 day training class presented by a guy named Henry Ott. The presentation was on the subject of grounding and shielding. I have the book from the class. A whole lot of practical explanations of things that make good or bad antennas. Skin effect, proper signal trace matching, terminating. pigtails, etc. Some of the reps I got to work with personally. I would have to look through business cards to recall their names. But their visit may have been arranged through Avnet or other distributors. The one TI /app engineer, helped me with the code for the Bluetooth key FOB way back in 2014 or so. I think we used code composer studio for that one. I did design work with the TI RFID glass device, and energizer detector circuits. Here in Florida I worked at Disney's Animal Kingdom for a year. The Kilimanjaro Safari vehicles use the same TI RDFID parts. I worked on the Rivers of Light attraction most of that year. Of course working at Cat I had mechanical guy, PCB layout guy, a power supply guy, a project lead, to deal with meetings and buying stuff. I took care of everything electronic. I captured the schematic, researched devices for power requirements, voltages, timing, temperature, availability etc. Needless to say I get to do all those tasks now for this design. I left Cat at the end of 2015 because they were having a bad year, and they offered me a buyout severance deal.

    By the way, I started using that new Brave browser. It seems that it blocks by default just about every website. This includes TI. That is why I couldn't find the specs the other day for the CODEC. 

    I just yesterday seriously thought that I could start with a more simple design before going for 8 channels. So I am thinking of using the cc8520 with a stereo CODEC supported by the PP tool, the TLV320AIC3204. That CODEC is not BGA, but some QFN package that I might be able to solder, along with the cc8520. Though the center pad could pose a problem on both devices. Perhaps I can have a slightly larger plated hole in the PCB for the pad, and flow solder through the hole. I have read of using a small toaster oven to solder parts like this at home. Just be careful not to over cook the devices. Perhaps I can layout a PCB in 2 layers using 0603 discrete parts. It may allow routing a trace through the center of the device. The TLV320AIC3204 seems to have differential line input, but single ended line output. Perhaps this idea could keep my expenses to less than $200 or so. Perhaps I could learn a whole bunch from this.

     I don't plan on the product being used within feet of routers and such. Cell phones with BT may be in the area, within feet.

    What state are you located in? Most of the Cat. reps/disty's were out of Chicago, or St. Louis.

  • Hi Paul,

    Thanks for sharing, I got a few years to catch up on compared to your experience.

    As for the design, while I think a 2 layer design is possible, it could be tricky to get a good layout done depending on the schematic layout. One if the things is that the recommendation is to filter both the VDD and VDDIO inputs to the device independently. This basically means you get 2-3 rails to route around, and not all of them is located conveniently close to each other. Long story short, the tricky part (if you ask me) of the 2 layer design is to get the power routing done without sacrificing optimal GND return paths and GND plane size in general.

    I would recommend going with a 4 layer design even in the simpler designed you might or might not be building.

    Also, if you would to consider using range extenders in addition to the CC85XX I would recommend looking at the CC2592 and not the CC2590/91 that is prompted in the tool. The "92" is a newer take on these and should work well with the CC8530 as well (using this in my private projects), it is absolutely easier to design with then the 90/91 version. As a "pro", you also get an easy differential to single ended conversion of the RF lines when using the range extender in addition to the CC85XX device.

    No matter if you decide to go with a smaller test design or not, I would recommend to see which ICs you might be able to sample cheaply from TI.com (if you have not checked this already):

    https://samples.ti.com/store/samples/en/?site=samples&clear=true

    I've done several private designs with the CC8530 devices myself, the overall design is rather simple to get up and running (I have however always used range extenders in addition to the CC8530). I'm mostly a hot-air gun type of person when it comes to the QFN packages (at home at least). I've been using the OSH Park service (an US fab) for the PCBs, I find these to be a good choose for small board prototyping, their 4-layer stack up is also reasonable priced for the specs you get (seen out of the eyes of a hobbiest builder).

    I have seen the different DIY "re-flow" ovens etc but as you can get a knock off brand SMD "rework hot-air gun station" super cheep now a days I typically recommend going with the hot-air approach if you do not feel like experimenting with the toaster oven.

    As for my location, I'm working out of the Oslo office in Norway. As for browsers now a days, that is basically becoming a science of its own..