That1580 mini mixer

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
mediatechnology
Posts: 5443
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: That1580/1510 mini mixer

Post by mediatechnology »

Run picoscope sign gen full scale (2v) to increase s/n ratio ?
Yes, exactly.
Already too hot for the mic input, but add in gain of 2 from the drv134 and it needs a pad.
Been saying you needed to pad the output for a number of posts.
This gives the PicoScope output its maximum S/N ratio.
When you pad the signal to get it to a usable mic level you also pad the generator noise with it.
600ohm output of the sig gen into 10k input on the drv134... not sure if this is much of a problem for basic testing?
When the DRV134/THAT1646 are fed by high impedance sources their noise level is higher.
The source impedance becomes part of the internal resistor networks and unbalances it.
The cross-coupled and OutSmarts® parts have high common mode noise gain to make them work like a transformer.
When an external resistor unbalances the network some of the common mode noise gets converted to differential.

But in your case, the added noise from a 600Ω source feeding your DRV134 will be less than what you have trying to use the PicoScope ARB generator's unpadded output.

Why don't you just build a U-pad using three resistors and be done with it?
Make the middle resistor 150Ω.
7K5Ω input legs will provide about -40dB attenuation.
I would be willing to bet if you had the PicoScope ARB output drive the hot input to the pad and grounded the other input leg it will work just as well as using the DRV134 to balance it.
Hook the DUT mic input's tip and ring to the 150Ω resistor.

The PicoScope will eventually limit you owing to its 12B A/D D/A.
But it does a lot of things really well.
weroflu
Posts: 56
Joined: Wed Apr 19, 2017 9:19 am

Re: That1580/1510 mini mixer

Post by weroflu »

I didn't mean to say that I was offering new information about the pad, I'm reading intently. It was sort of a mis-edited post that came across like I came up with the idea. I'm still trying to wrap my head around some concepts but they are slowly sinking into my thick skull.

The pico is not even 12bit, it's 8bit with a 12bit software oversampler that I don't really understand.

Even though it seems like induced noise by the drv134 will not be an issue for my purposes, do you think using the LL1527 balancing transformer has any benefits for the same purpose? Along similar lines do you think it's worth it to get a proper signal generator?

Hit a milestone with this project yesterday. I recorded a stereo pair of mics output through the master section and the whole signal chain worked. Two audible issues are clicking and maybe some sort of loose connection where the sound got louder/fuller about 1 minute in.
User avatar
mediatechnology
Posts: 5443
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: That1580/1510 mini mixer

Post by mediatechnology »

Hit a milestone with this project yesterday. I recorded a stereo pair of mics output through the master section and the whole signal chain worked. Two audible issues are clicking and maybe some sort of loose connection where the sound got louder/fuller about 1 minute in.
Cool!
I didn't mean to say that I was offering new information about the pad, I'm reading intently.
I didn't take it that way and thank you for reading intently.
I started suggesting the use of a pad way back here: viewtopic.php?f=6&t=882&start=4
It was simple enough to try I was hoping that you would have already tried it.

I would save your money on the LL1527.
I think that with a U-pad on the OP of either the PicoScope or DRV134 you can get a good enough signal for troubleshooting.

Since the PicoScope ARB Out has a 600Ω impedance you'll want to add about 600Ω to the bottom grounded leg if it drives the pad directly.
This makes the Picoscope Output "impedance-balanced."
If you use the DRV134 to drive the pad it's already 25Ω per leg and you don't need to add a resistor to compensate the PicoScope's Rout. (I know the DRV134 looks like 50Ω Rout but isn't.)

For the pad.
The top leg (Picoscope output or DRV134 +) would have 7K5 series feeding a 150Ω shunt.
The bottom leg (PicoScope ground or DRV134 -) should have 7K5 (+600Ω if not DRV134) feeding the bottom of the 150Ω shunt.
The 150Ω shunt feeds the mic input + and -.

If you don't have 7K5 then use 4K99, 10K or whatever you have close.
The attenuation won't be precise until you calculate it based on what resistor value you have and you can adjust using the DUT gain or PicoScope ARB output level.
If you don't have the added 600Ω resistor it will still probably work but may show a little more hum.

You won't be able to make noise or distortion measurements with the PicoScope.
It's just doesn't have the dynamic range.
You need a good sound card or outboard converters.
weroflu
Posts: 56
Joined: Wed Apr 19, 2017 9:19 am

Re: That1580/1510 mini mixer

Post by weroflu »

I noticed the 5171 and 1580 seemed to be a little too warm. With no load they are peaking at 50c and 55c respectivelty, after about 5-6 minutes.
User avatar
mediatechnology
Posts: 5443
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: That1580/1510 mini mixer

Post by mediatechnology »

They do run fairly warm to the touch.
weroflu
Posts: 56
Joined: Wed Apr 19, 2017 9:19 am

Re: That1580/1510 mini mixer

Post by weroflu »

Update

Two channels now working pefectly through the whole signal chain, preamplifier, spi gain set, pan and sum, no noise.

edit: the clicking was caused by something on the summing card. I reseated or replaced, can't remember, the lme49720 and that did it.

I have tentatively decided to stay with the current design for this revision since things are working and the 1206/1286's are already purchased. Next revision, if it happens, will take all of the suggestions in this thread into account, maybe look into cmos switches, metering or headphone amp. etc (it would be something to squeeze in that pcm4222 from ther other thread). Also I realized that I might be able to eliminate not just the 1286 but also the 1206/1246 if summing was balanced. Now that I hear no noise in the chain with everything going, including bluetooth turned on, more seems possible.

I can't remember if I used a 5171 from the graveyard on the 3rd preamp card, but I finished soldering the 3rd one. 5171 spi worked fine. FIred up the card and no signal.
Fluxed the edges of the 1580 and ran iron around it to no avail. Replaced the 1580, didn't help.

Fluxed around the 5171 edges and touched up with iron and got some loud distortion with signal on top.

Maybe I should either try to hot air reflow that 5171 or replace it.

The 1580's now perfectly self-center. I went back to using the technique of applying iron to bottom side exposed pad and it centers very fast.

I have no idea if it's the 1580 or 5171 causing the problem, but it seems like one of those two is the culprit. Never got any of the 5171's to self center.
If It turns out that is the problem then I might pick up a board preheater and use that for the underside and hot air on the top.

Working on getting audiotester to recognize my soundcard under wine. It seems possible since the sound card is recognized running audacity under wine.
weroflu
Posts: 56
Joined: Wed Apr 19, 2017 9:19 am

Re: That1580/1510 mini mixer

Post by weroflu »

Nope spoke too soon.

Very low level rhythmic clicking audible if I turn up the headphone volume. 3-4 per second.
Forgot I enable bluetooth broadcast in preparation for the next round of coding torture.

RFduinoBLE.advertisementInterval = 250

hmm, 250ms x 4 sounds about right. Disabled bluetooth broadcast and clicking stopped.

This is good it will force me to start on the bluetooth coding. The backup plan for which is to stick a mini usb connector on this thing and do cabled gain setting.
Also I've been testing everything with the pcb unshielded, maybe it's good to put it in a shielded box?
User avatar
mediatechnology
Posts: 5443
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: That1580/1510 mini mixer

Post by mediatechnology »

Shielding is a good thing.

There might be RF rectification of the Bluetooth RF envelope.
Do you see anything on the power rails?
weroflu
Posts: 56
Joined: Wed Apr 19, 2017 9:19 am

Re: That1580/1510 mini mixer

Post by weroflu »

Haven't yet checked the power rails. I got stuck on these qfn32's again. FInally got one more done after a zillion tries, either not enough solder on pads, too much on the thermal or something else. Card #4 has the 5171 on, it seems the rest of the populating is fairly easy once that's done. Card #5 has the 5171 on but some weird behavior. When I test it by printing the gain register value in serial console the first 3 times I run the test sketch everything looks good. Then from the 4th time on the gain value is correct but it shows as channel 0... ??? I triple checked and replaced the gpio address set resistors but it didn't help. Maybe the chip is not seated properly still. Or maybe there is something wrong with my layout involving this particular resistor since I've not gotten to bit 3 yet. Will try to set address as 5,6,7 to see if they all give me the same problem.

Also looking into a backup plan instead of bluetooth using a usb to uart bridge chip on the rfduino adapter board or just a usb/uart cable, for a wired connection from phone to the rfduino. I might pursue both simultaneously and the best horse wins.
User avatar
mediatechnology
Posts: 5443
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: That1580/1510 mini mixer

Post by mediatechnology »

I'm glad I'm not the guy soldering this.
I'm amazed.
Post Reply