About 3 days ago, I found out that the dir1703 was pulled out of TI's web site and I was wondering if anyone here knows what's happening. I emailed TI and haven't heard from them. Several forum members have expressed difficulty in getting this chip to work properly, and I am just curious if TI has decided to pull this chip off the database because of some design flaws. I was originally planning on using this chip in the 44.1KHz fs mode but decided against it since no one has proven that the chip works in this mode.
For those that have used this chip, what are your impressions?
Thanks,
rlim
For those that have used this chip, what are your impressions?
Thanks,
rlim
Don't know what happened, seems to have disappeared from TI-site, although samples still available according to TI-site!!!!
Arnaud
Arnaud
The part is still mentioned in many current documents from TI. For example, there is the Audio Solutions Guide, which contains DIR1703 pricing information and a link to the product folder.
But it definitely vanished from all databases and also the direct link http://www.ti.com/sc/device/DEVICENAME points to an error page.
There was a similar issue with the AD1896 ASRC. The samples link disappeared from the Analog Devices page without notice. When they fixed some problems with the device, a new AD1896A version appeared and was shipped instead.
Maybe it's just TI's way of admitting there is a problem. Has anybody of you ever seen the DIR1703 in a commercial product anyway?
But it definitely vanished from all databases and also the direct link http://www.ti.com/sc/device/DEVICENAME points to an error page.
There was a similar issue with the AD1896 ASRC. The samples link disappeared from the Analog Devices page without notice. When they fixed some problems with the device, a new AD1896A version appeared and was shipped instead.
Maybe it's just TI's way of admitting there is a problem. Has anybody of you ever seen the DIR1703 in a commercial product anyway?
Oh my god!! I'm using DIR1701 in my own DAC.....
Is there any simple cure?
How about frequency to voltage converter followed by differentiater, to generate the reset pulse in case of changes in sampling rate?? Any idea?
Is there any simple cure?
How about frequency to voltage converter followed by differentiater, to generate the reset pulse in case of changes in sampling rate?? Any idea?
This might be why I have a few problems with mine, I cannot lock onto anything other then a 48khz sample frequency. Also there are some performance issues I have that could be related to the DIR1703. A replacement with the same pin layout would be good.
5th Element
Does it work if you start sampling at 44.1KHz after power on? Seems like you shouldn't get any problems if there is no resampling of other fs.
rlim
Does it work if you start sampling at 44.1KHz after power on? Seems like you shouldn't get any problems if there is no resampling of other fs.
rlim
An externally hosted image should be here but it was not working when we last tested it.
I use DIR1701 and it has no problem working with 44.1KHz input.
I remember when I was doing initial test with my DAC, it didn't have problem when I interchanged the input rate between 44.1 and 48KHz. But I don't know about other sampling rate, or whether the problem appears in individual chip but not all...
Re: 5th Element
Im not sure I have not tried it. I can set the master clock to whatever I like 128,256,512fs and it only detects a 48khz when an input is applied. Brate0 and 1 change output state depending on what "fs" is being used. I thought they were just meant to change to show what sample rate is detected by the DIR but no luck.
BRSEL is open and I use a 12.288Mghz crystal set at 256fs (SCF0 high, SCF1 low) which denotes 48khz sample rate. But if I change the clock to 512fs the brate pins state change but I can still only get a sample rate of 48khz to work.
Interesting looking at the data sheet is looks as if you cannot get 44.1 and 48 to work of the same freq crystal? maybe this is the problem. CKSEL is grounded.
rlim said:
Im not sure I have not tried it. I can set the master clock to whatever I like 128,256,512fs and it only detects a 48khz when an input is applied. Brate0 and 1 change output state depending on what "fs" is being used. I thought they were just meant to change to show what sample rate is detected by the DIR but no luck.
BRSEL is open and I use a 12.288Mghz crystal set at 256fs (SCF0 high, SCF1 low) which denotes 48khz sample rate. But if I change the clock to 512fs the brate pins state change but I can still only get a sample rate of 48khz to work.
Interesting looking at the data sheet is looks as if you cannot get 44.1 and 48 to work of the same freq crystal? maybe this is the problem. CKSEL is grounded.
But isn't that exactly the difference between dir 1701 and 1703? 1701 can work with different samplingrates on the same crystal frequency, whereas the 1703 does need different crystals for different samplin rates? I always wandered what was the point in the 1703 then, but maybe it has better jitter specs?
If that is so then thats why mine doesnt work how I thought it should haha. I just saw that everyone used the DIR1703 so went with that one.
Dear Rlim, I too found the DIR1703 to be deleted from TIs site several months ago. A quick check with the US vendors showed no stock avalability.
But, a few nights ago, I found the vendor one again had stock of over 1K pieces. I contacted TI to see what was up as I inted using this devise in comercial applications. TIs reply reads quote'
Unfortunately, we have no replacement models for DIR1703/01 at present.
Therefore, CS8414 or CS8416 from Cirrus Logic are recommended.
Seems sad to me...
But, a few nights ago, I found the vendor one again had stock of over 1K pieces. I contacted TI to see what was up as I inted using this devise in comercial applications. TIs reply reads quote'
Unfortunately, we have no replacement models for DIR1703/01 at present.
Therefore, CS8414 or CS8416 from Cirrus Logic are recommended.
Seems sad to me...
- Status
- Not open for further replies.