and now for something completely different part 3

[mwl168]

deleted - merged with previous post

 

Edited November 16, 2017 by mwl168

[Pars]

Mouser is pissing me off... placed an order on Sunday night. It didn’t ship until late Tuesday, using that UPS Innovations crap. Supposedly in town being transferred to USPS, but don’t know what’s up with it or when it will be delivered. I was hoping to get the CFP boards at least in a testable state this weekend. Fuckers...


Sent from my iPhone using Tapatalk

[mypasswordis]

I've always used standard priority mail with mouser and never had any problems with that. They do take two days to ship, but at least it's consistent.

[dsavitsk]

Mouser shipping has had issues for a few years now. Packages rarely seem to ship on the day they are supposed to. That said, they always upgrade me to 2 day, or sometimes overnight, to get them here on time.

[mypasswordis]

Weird, maybe I just get lucky as it's like clockwork for me. Two days to ship, two days shipping, since my very first order or at least as far as I can remember.  

[Tinkerer]

58 minutes ago, Pars said:

Mouser is pissing me off... placed an order on Sunday night. It didn’t ship until late Tuesday, using that UPS Innovations crap. Supposedly in town being transferred to USPS, but don’t know what’s up with it or when it will be delivered. I was hoping to get the CFP boards at least in a testable state this weekend. Fuckers...


Sent from my iPhone using Tapatalk

I've been using Arrow. They usually ship next day and give you free overnight if you spend more than $20 on an order. And regular shipping is still free on lesser stuff which makes it ideal when I'm repairing something and just need a couple bucks worth of stuff.

Edited November 17, 2017 by Tinkerer

[Pars]

Yes, I’ve used Arrow in the past year for some things, as well as OnlineComponents, though they have a $50 min order. They also have some Pana TS series caps, which makes it worthwhile
Got an update from USPS... order out for delivery. This is the first time I’ve had trouble with Mouser... their Econ shipping normally arrives in 2 days.


Sent from my iPhone using Tapatalk

[mwl168]

Got the XLR socket in. Here is the finished amp. The volume pot is cosmetic only - not wired.

 

Edited November 18, 2017 by mwl168

[n_maher]

Elegant work, but if I may, that chassis begs for the XLR socket to be mounted behind the front panel to maintain the otherwise clean aesthetic.  I'd also recommend removing the locking tab, I never had much use for them, unless it's catastrophic to unplug something while the amp is on.

[mwl168]

9 minutes ago, n_maher said:

Elegant work, but if I may, that chassis begs for the XLR socket to be mounted behind the front panel to maintain the otherwise clean aesthetic.  I'd also recommend removing the locking tab, I never had much use for them, unless it's catastrophic to unplug something while the amp is on.

Thanks Nate. 

On the hindsight, I wish I had ordered the XLR socket in the silver color. May still give that a try.

 

[n_maher]

The silver XLR would also be a nice option and require significantly less work.  To rear mount the socket, given how thick the panel is, you'd have to recess the body of the socket into the panel which is no trivial exercise - unless you're Kerry and have a cnc router in your closet.  

[mwl168]

Exactly my thought Nate. Mounting behind the panel will require some recess on the back of the panel.  I checked my closet but did not find a CNC router in there.

Plus, it will still have the two holes for the screws exposed on the front panel which is not ideal either. It would be great to have XLR jacks made similar to Justin's Stax socket.

Never scored high in the art classes I took. The black XLR socket looks fine on my silver SS Dynalo chassis but somehow looks a bit off on this particular chassis. Maybe I should put a black color knob on it...

 

 

[mypasswordis]

Weird that I looked and found one in mine. the front panel is thick enough that you can potentially blind drill and tap. Pm me if you still want this done in January, I'll be out of town most of December. 

[n_maher]

The screws can be dealt with, you just counter sink them so they sit flush (since you already drilled the holes) or use a button head screw to keep the profile as low as possible.  

[mwl168]

Thanks guys. Just ordered the silver XLR socket and let's see how that works. 

I want to keep Nate happy!

[Helium]

After mwl168 wrote his impressions on the sound, this amp looks very temptating.

Ordered 20pcs PCB (SE version, so 4 needed for balanced bridged amp). Plan to use Toshiba BJT as output transistors, as I was unlucky with matching MJE15030 and 15031. Never tried to match MJF though, but I think they are hardly different from MJE.

[kevin gilmore]

Make sure the toshiba are the matching pinout

[Pars]

3 hours ago, Helium said:

After mwl168 wrote his impressions on the sound, this amp looks very temptating.

Ordered 20pcs PCB (SE version, so 4 needed for balanced bridged amp). Plan to use Toshiba BJT as output transistors, as I was unlucky with matching MJE15030 and 15031. Never tried to match MJF though, but I think they are hardly different from MJE.

Which Toshibas? And it's tempting, BTW

I'm just going to use my cheapie meter with hFe to see what the MJE1503xs look like instead of rigging up a jig and doing it properly.

[Helium]

Yeah, English is not my native language

I'm going to use Toshiba 2SC5171/2SA1930 as output BJT. Pinout is same to MJE/MJF.

Regarding matching, I'm trying to figure out what Ic or Ib or Ie I should aim at while matching them for CFA circuit.

I've got different devices for measuring hfe. One of them declares measuring hfe @ Ic=5mA, others do this @ Ib=10uA, 1mA, 10mA. So I'm a little bit stuck here (Ic vs Ib, and appropriate range).

mwl168  mentioned he's running 220mA bias current. Can it be asumed that 220mA is Ie? Maybe Kevin could chime in?

[kevin gilmore]

the pair of output transistors run at about 200ma

the drivers at about 20ma

vbe transistor runs at about 10ma

 

[mwl168]

4 hours ago, Helium said:

 

mwl168  mentioned he's running 220mA bias current. Can it be assumed that 220mA is Ie? Maybe Kevin could chime in?

I started running mine at 150mA and experimented with higher bias up to 220mA and settled for 200mA. Also experimented with supply rails from 22VDC to 26VDC. Also tried with and without the servo engaged.

Frankly, in the balanced mode, I did not hear a big difference. YMMV.

With servo in, the DC offset of each channel and each board is within 1mV according to my DMM. Without the servo, DC offset on each board sits around 60mV and does not change much as the amp warms up, and in balanced mode, the offset is about 30mV and very stable. I did not match the MJF output devices.

   

[Pars]

4 hours ago, mwl168 said:

I started running mine at 150mA and experimented with higher bias up to 220mA and settled for 200mA. Also experimented with supply rails from 22VDC to 26VDC. Also tried with and without the servo engaged.

Frankly, in the balanced mode, I did not hear a big difference. YMMV.

With servo in, the DC offset of each channel and each board is within 1mV according to my DMM. Without the servo, DC offset on each board sits around 60mV and does not change much as the amp warms up, and in balanced mode, the offset is about 30mV and very stable. I did not match the MJF output devices.

   

Crap. Just noticed I ordered MJEs, not MJFs...

[Helium]

8 hours ago, kevin gilmore said:

the pair of output transistors run at about 200ma

 

Kevin, so is matching them at Ib=1mA OK? (provided that hfe is ~200)

1 hour ago, Pars said:

Crap. Just noticed I ordered MJEs, not MJFs...

Why? They are basically identical, except that MJE can sustain more power (=better), but doesn't have isolated tab.

[Pars]

The isolated tab. I didn't know if I had enough to220 thermal pads and shoulder washers around... I do.

EDIT: Ack, Mouser. The 5K trimpots I ordered (and ordered correctly) turned out to be 25K. Label on bag was correct for 5K. Must have been a good day in order picking . They also shorted me some 560 ohm resistors, but thoughtfully provided the 4 I needed.

Edit 2: I put 10K pots in with a 10K resistor in parallel. Will change to the 5K pots when Mouser fixes it.

Edited November 20, 2017 by Pars

[Pars]

Fired board #1 up tonight on a bench supply at +/-20V. Seems to be working, characteristic 58mV DC offset. At first I wasn't seeing anything across the 1R resistors, but after turning the pot clockwise, it started going up. I left it at around 158mV, which should be 150mA bias? I didn't feel any heat from the transistors, and I don't have the L bracket bolted onto a heatsink yet, so not really interested in cranking it up much. Also didn't leave it on long at all, less than 2-3 minutes.