stef1777 said:
Yes they are normally hot to the touch. They are so small and they pass IIRC 40mA. What's your supply voltage?
Jan Didden
janneman said:
Hi Jan!
+-15v for power supply and 2.5v for Vocm. I changed R9 for 39K to get 2.5v. 62K was too high (2.4v only).
http://www.awdiy.com/uploads/pdf/DEQminiAnalog-beta14.pdf
I also changed the opa1632 gain for 1.6 (R1 = 1K8, R3 = 2K7). This show same level value on the RTA as using regular XLR input. I've to find more resistors to try an higher value on the feedback resistor. 1K8 is pretty low at input (but should be fine with my TAG).
Stef...
stef1777 said:
I run mine at +/- 9V and they still get quite hot. Seems normal for these things.
At 35 degr ambient you can dissipate around .7W in the SOP package. At 14mA supply (not 40mA I said earlier!) and +/-15V that's .42W, on the safe side. Only at 60 degr ambient would you get to the limit. It's all detailed in the data sheet.
Jan Didden
on the analog input schematic of the DCX there is a high-pass shelving filter before the first OP (differential reciever ) and the two later ones that converts the single-ended to a differential drive which gives 6 dB higher gain at the higher end of audio spectra. (to improve SNR I guess)
Dont you need that aswell in this analog input board?
Or is the idea to always have a digital filter enabled with the same parameters regardless of what filter configuration one might use?
Or is that option to remove in the DCX ?
regards,
Rickard
Dont you need that aswell in this analog input board?
Or is the idea to always have a digital filter enabled with the same parameters regardless of what filter configuration one might use?
Or is that option to remove in the DCX ?
regards,
Rickard
rikkitikkitavi said:
I don't really know the DCX but on the DEQ, the "digital" filter is close to the ADC chip on the DSP board. For this reason, I've not put a filter in the input board.
Jan : I've modified the output board for more level. Input, output level is now matched (aka if I switch off the DEQ, level stay the same as original circuit). The filter is a little more noisy (on the simul) but acceptable. The best will be to have done a 2 stages output board.
No DC at output at all with the OPA2132.
1632 are at 58°C after half an hour. Pretty high. PowerPAD version will be better but so difficult to solder...
Stef...
Working!!!
Q=1 and 1vrms = -10dB on the RTA screen.
.
Q=1 and 1vrms = -10dB on the RTA screen.
An externally hosted image should be here but it was not working when we last tested it.
.
rikkitikkitavi said:
Yes I have noted that before, IIRC there is a complement in the output stages. I guess the idea is like they do for some CD's: a kind of pre- and de-emphasis, presumably to improve S/N ratio.
Most people remove the input circuits together with the output circuits so it doesn't matter. It *does* matter when you only replace the output stuff and not the input stuff or vice versa.
Jan Didden
stef1777 said:
Of course there is no DC 😉
58degr is OK, it's just we humans are sissies with heat, silicon is not 😉
Well done.
Jan Didden
janneman said:
I hate heat. 😉
One question Jan. For the OPA1632, I need Q=1.65. Curently, I use 1K8 at input and 2K7 for the feedback. What do you think to push the feedback resistor to 4K or 5K to have more inpedance at input?
http://www.awdiy.com/uploads/pdf/DEQminiAnalog-beta15.pdf
I'm trying different op-amp on the output board. Very funy.
Stef...
awright, janneman.
makes sense, but I thought the output filter was operating at much higher frequency to remove spurious effects from teh oversampling DAC (typically it is a bessel response , cutoff > 40 kHz)
as in a typicall applicaton for DACs.
will have to check schematics of the DCX again, I havent made any calculations of this.
makes sense, but I thought the output filter was operating at much higher frequency to remove spurious effects from teh oversampling DAC (typically it is a bessel response , cutoff > 40 kHz)
as in a typicall applicaton for DACs.
will have to check schematics of the DCX again, I havent made any calculations of this.
rikkitikkitavi said:
No, the cutoff is lower, I made some sims of the stock circuit and my passive output board that was published in AX.
But there was something in the input that wasn't straight, I'll try to find it back, must have it somewhere.
Jan Didden
stef1777 said:
The only issue with higher resistance would be increase of noise, but since the circuit will allways work (near) max input level anyway, it's really a non-issue.
Jan Didden
Hi Folks!
I made some listening session yesterday with some friends. The boards are a real improvement compared to the stock Behringer i/o board (not modified). If we compare my boards with a modified Behringer i/o board (caps and op-amps), the difference is difficult to heard.
The main problem for the input board stay OPA1632 temperatures. After 2 hours, box closed, I found the chips at 75°C and a little more, very very too hot. I had offset drift and bizarre low noise sounds (without sound input). I don't know yet how to solve this. I can try to lower the PSU but need to add vregs but this will use more parts and the vregs will also dissipate temperature. I can try to use the PowerPAD version but not sure of the result and they are very difficult to solder. Or use someone else of the OPA1632 but this is another story.
Stef...
Not happy.
I made some listening session yesterday with some friends. The boards are a real improvement compared to the stock Behringer i/o board (not modified). If we compare my boards with a modified Behringer i/o board (caps and op-amps), the difference is difficult to heard.
The main problem for the input board stay OPA1632 temperatures. After 2 hours, box closed, I found the chips at 75°C and a little more, very very too hot. I had offset drift and bizarre low noise sounds (without sound input). I don't know yet how to solve this. I can try to lower the PSU but need to add vregs but this will use more parts and the vregs will also dissipate temperature. I can try to use the PowerPAD version but not sure of the result and they are very difficult to solder. Or use someone else of the OPA1632 but this is another story.
Stef...
Not happy.
Stef, I don't see any problem with heat
I solved that with coolers and unit is reasonably hot now. Regarding decreasing voltage did you tried passive resistors? With adding come capacitance after it will do some nice R-C filtering and improve decoupling from the rest of active analog parts. And about that "bizarre low noise without input", did you tried to terminate the inputs with some resistance? Anyway , count me for one PCB, I would like to rid of opa's2134 driving 600ohm😉
I solved that with coolers and unit is reasonably hot now. Regarding decreasing voltage did you tried passive resistors? With adding come capacitance after it will do some nice R-C filtering and improve decoupling from the rest of active analog parts. And about that "bizarre low noise without input", did you tried to terminate the inputs with some resistance? Anyway , count me for one PCB, I would like to rid of opa's2134 driving 600ohm😉Moby said:Stef, I don't see any problem with heat
II prefer to restart the PCBs from scratch. The new versions will be vertical. This will solve heat coming from the AKM chips and space problems. The output board will fit on the side with a 2x6 pins connector.
New input board draft PCB.
An externally hosted image should be here but it was not working when we last tested it.
Connectors placement test. We can close the box (2mm left at top).
An externally hosted image should be here but it was not working when we last tested it.
stef1777 said:
Steff,
I think the OPA1632's are fine at 75degrees. If you feed them from +/- 15V, they dissipate about 450mW. They are designed for that. Even if you cannot touch them, 75degr is not hot by ss standards. With 450mW and at 170degree/Watt thermal resistance, the die will be about 76degr above ambient. If the case air temp is 40degr, the chip will be internally, the die, at 40+76 is 116 degr. The max is 150degree, with recommended operating max of 125degree. So, no problem.
You should stop thinking about this with your 'human' feelings of heat 😉
Jan Didden
janneman said:
The OPA1632 is not the only reason to decide to do a complete new board.
On the current board, the OPA1632 are just at the top of the AKM chips. The ADC on the right is the hottest chip. Vents is not good at all. I have 10/20°C more with the analog board (box open). Box closed it should be worst. Caps goes very hot too. The original caps are more high too and the analog board can't be plugged.
I also have a "mechanical" problem with the output board. The board don't fit at top and need to be completely reworked.
The parasite sound I got after half an hour (box closed) seems to come from the OPA1632 or someting close when too hot. I don't have this parasite without the board. I used a freeze spray on the OPA and the sound stopped. And as wrote in the datasheet, the OPA1632 are not thermal protected...
It's not a problem. I just lost times and the cost of the pcb and connectors. It's just a pain to redo from scratch a work already done.
First numbers without board, second with analog addon board. Box open.
An externally hosted image should be here but it was not working when we last tested it.
Stef, I think that moving pcb around will not help much because opa1632 and all of the chips are heating the air inside the box. I don't understand why you say "vents are not good at all"?? I was thinking about fan coolers and maybe to open few more holes at the sides of the box because with just top hole you can't have air circulation...
Moby said:
With a large window holes at the top, it's currently ok with the diy PSU. Nothing is more that 60°C. Adding some holes on the sides is better as well. Don't add holes under the box because heat coming from others equipments placed under the DEQ will add more heat.
Moving the analog board vertically will be better as heat will not be trapped between the DSP board and the analog board. Heat coming from the AKM chips will no longer hit the analog board directly adding heat to the OPA1632.
If the analog board is powered and placed vertically, the OPA1632 stay around 54°C.
Stef...