Thank you Brian, I did reset (with power on), and started with following tests from manual in figuring out the problem. I will let you know once I confirm the problematic board. Reset did not help. I think I remember that when I purchased his unit from one really nice guy, who was moving from San Francisco to Boston that he mentioned he had some problems with Program RAM and that he fixed it. It seems like I will be going with the same path.
Sorry Tom that I hijacked your thread.
Thank you very much. I am member as well. Let me find time to do first my due diligence. Service manual is very helpful and detailed. I will definitely ask questions once I hit a stumbling block on my own. It is not nice to ask dumb questions without doing all I can on my own.
Happy New Year to you sir as well!
I am considering your driver to use for converting the SE output of my Sony 9000Es to balanced to feed my balanced preamp. The CD player has an impedance of 220R and the preamp has a buffer input.
Do I still need to populate the buffer section in your driver? Will sound quality be better with it? The Driver will be placed right next to and in between the CD player and preamp with 9 in cables.
Are there any DC offset issues with the Driver and receiver to be concerned about?
thanks nash
Do I still need to populate the buffer section in your driver? Will sound quality be better with it? The Driver will be placed right next to and in between the CD player and preamp with 9 in cables.
Are there any DC offset issues with the Driver and receiver to be concerned about?
thanks nash
I'd still use the buffer op-amp. The input impedance of the THAT1646 is pretty low. 5 kΩ if memory serves.
Nope.
Tom
Thanks for clarifying.
Perhaps not a valid question for here but how do these THAT 1646 drivers compare sonically to a transformer like the Jensen JT11P1?
nash
The JT11P1 is an input transformer. The THAT1646 is an output driver.
The Jensen transformer makes a dull thunk when I drop it on my workbench. With the THAT1646, it's more a tick sound.
The transformers generate much more THD than the IC drivers and receivers.
Tom
Hi, have you been back in the lab yet? My guess is it will measure the output noise close to -101dBu which is what THAT specifies in their datasheet.
Oh, I didn't realize you wanted me to measure that. I offered to measure and never heard back from you, so I moved on. Sorry...
I don't have a fully assembled THAT TX in stock, so I can't grab one to measure, unfortunately. My DIFF PRE 8x2 measures >112 dB SNR. That's the measurement limit by the AP. The DIFF PRE 8x2 is essentially four THAT RX -> relays -> Alps RK27 volume pot -> THAT TX. The THAT TX should be a touch better as it should have a slightly lower noise floor. But as the DIFF PRE already hits the limits of the APx525, I won't be able to measure the THAT TX if it is better than the DIFF PRE.
Tom
Since you have a APx525, measuring the THAT TX is very easy!
Short the input to the THAT TX, connect the output directly to the APx input and simply measure the RMS level with a un-weighted 20kHz measurement bandwidth. This gives you a single real value for the output noise level.
The APx525 is specified with 1.3uV (-117dBV) input noise, so only if you measure close to that can we conclude that the THAT TX has less output noise than your APx input noise.
Where do you get the 112dB SNR measurement limits of the APx from? I can't find it in the specs. It can measure from 1.3uV to 160V...
Short the input to the THAT TX, connect the output directly to the APx input and simply measure the RMS level with a un-weighted 20kHz measurement bandwidth. This gives you a single real value for the output noise level.
The APx525 is specified with 1.3uV (-117dBV) input noise, so only if you measure close to that can we conclude that the THAT TX has less output noise than your APx input noise.
Where do you get the 112dB SNR measurement limits of the APx from? I can't find it in the specs. It can measure from 1.3uV to 160V...
Yes. Very easy if I had an assembled board.
As I said earlier, I do not have an assembled board in stock so I do not have one handy to measure. That makes it more challenging and more expensive to conduct the measurement. Thus, I point you to my DIFF PRE 8x2, which measures 21 uV (20 Hz - 20 kHz, unweighted) and consists of THAT RX -> relay -> 100 kΩ Alps RK27 -> THAT TX. It follows that the noise floor of the THAT TX is no higher than this value, no?
Next time I get an order in for a fully assembled THAT TX, I'll be happy to measure its noise floor for you.
I would also appreciate it if it's the SNR you're after or just the integrated noise.
Tom
To give you a bit better precision on the noise estimate: The THAT TX is a THAT1646 and an LME49710. The LME49710's noise integrated over 20 Hz to 20 kHz is 0.35 µV RMS. The THAT1646 is specified to -101 dBu (6.9 µV RMS). Even when taking the 6 dB gain of the THAT1646 into account, the LME49710 is well below the noise floor of the THAT1646, hence, the output noise of the THAT TX board will be dominated by that of the THAT1646.
In my experience, THAT is honest with their specs. The parameters I have measured on the circuits I've built with the THAT parts have all met the data sheet performance. I fully expect that to be the case for the THAT1646 as well and will be happy to verify once I get another order for a fully assembled board.
Tom
In my experience, THAT is honest with their specs. The parameters I have measured on the circuits I've built with the THAT parts have all met the data sheet performance. I fully expect that to be the case for the THAT1646 as well and will be happy to verify once I get another order for a fully assembled board.
Tom
-152 dBV is the noise floorof the FFT. You can see it here (and on my website):
There is a catch: The FFT bin width impacts the noise level measured. Using the FFT noise level rather than the integrated noise level was a poor choice on my part. I'm sorry. For future measurements I will use the integrated noise (unweighted and A-weighted).
Tom
An externally hosted image should be here but it was not working when we last tested it.
There is a catch: The FFT bin width impacts the noise level measured. Using the FFT noise level rather than the integrated noise level was a poor choice on my part. I'm sorry. For future measurements I will use the integrated noise (unweighted and A-weighted).
Tom
Sounds like a great question to ask your AP rep. My comment referred specifically to the Jensen JT-11P-1. You can see the full context around Post #69.
Tom
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.