You mean this.. (attached)
I looked up these "large black cap" devices that were already
soldered across the relay contacts. They crossed over to a
"Auftrag CURRENT LIMITR INRSH 2.5 OHM 10A-SL22 2R510 "
Are these the devices you speak of ?
The original 2 - 1.8KVA each had a independant relay / inrush
mov and were fed by a 10A fuse on the back panel.
Thank you andy.. I might just visit the PS forum now that I
have one that scares me...
OS
Attachments
soft start on 110/120Vac
Hi,
yes that 4off 22r plus the DC resistance of the primary windings limits the peak start up current.
If you had 120Vac and primary=1r0, then the start up currents peaks at 120/[22+1]*1.414<=7.4Apk This lasts for a few fractional of a ms and falls rapidly to the working current in the transformer.
Set that relay delay to ~ 200ms to 300ms.
I don't know what the 2r5 etc are describing. They could be the cold resistance of a PTC Thermistor, but that's a guess.
The SL22 could lead to a datasheet.
Hi,
yes that 4off 22r plus the DC resistance of the primary windings limits the peak start up current.
If you had 120Vac and primary=1r0, then the start up currents peaks at 120/[22+1]*1.414<=7.4Apk This lasts for a few fractional of a ms and falls rapidly to the working current in the transformer.
Set that relay delay to ~ 200ms to 300ms.
I don't know what the 2r5 etc are describing. They could be the cold resistance of a PTC Thermistor, but that's a guess.
The SL22 could lead to a datasheet.
You were on the money.. sl22 returned this..
http://www.ametherm.com/Data%20Sheets/SL22%200R712.pdf
12 Amp thermister .. they had this in parallel with contacts,
no power resistors , relay activates after the CPU verified no shorted amps.(original circuit) Did they use this device
instead of the resistors ? Would the resistors be better?
OS
http://www.ametherm.com/Data%20Sheets/SL22%200R712.pdf
12 Amp thermister .. they had this in parallel with contacts,
no power resistors , relay activates after the CPU verified no shorted amps.(original circuit) Did they use this device
instead of the resistors ? Would the resistors be better?
OS
I guess, for thermistors, they would taper the inrush rather than just hard limit the current until the relay closes, which I guess is less stressful. They're also desigend to run hot.
I guess resistors have the advantage that finding suitable power resistors is easy - personally I prefer a 20W aluminium clad type.
edit: dumb typo fixed
I guess resistors have the advantage that finding suitable power resistors is easy - personally I prefer a 20W aluminium clad type.
edit: dumb typo fixed
Member
Joined 2009
Paid Member
Soft start
Below is the one I'm using. Thermistors have the advantage over resistors in that they can withstand the current through if the relay doesn't close - resistors will NOT.
I use the power supply positive leg to power the relay. The series resistors and the zener cut the Vcc down to 12 VDC.
Below is the one I'm using. Thermistors have the advantage over resistors in that they can withstand the current through if the relay doesn't close - resistors will NOT.
I use the power supply positive leg to power the relay. The series resistors and the zener cut the Vcc down to 12 VDC.
Attachments
Because Class AB amps do not draw a constant current.
Degaussing uses a thermistor to introduce a taper off in the current gradually, they are PTC's and not NTC's. Incidentally try this on a monitor, degauss a few times in a row. You will notice the 2nd and 3rd degausses dont work until the internal thermistor has cooled.
I already have 2 of the thermisters , mouser and digikey have them as well. You are saying that all I have to do is delay the
relay unit until the thermister's current fully energizes the supply caps ?
(attached)
I want to go "dummy" (simple), until I add a remote controlled
preamp to this project (2 lm3886 rear amps as well , as I have
a 300VA 30-0-30 to use as well). Then I might play with some
of the logic controlled soft starts andy linked to.(fancy on/off
switches)
Edit:
I was searching your thread for that this morn.. thanks.
I won't just use the MOV by itself , I reserve that BS design
mentality for walmart junk. (I'm sure some would do it , but
not me.)
OS
Attachments
Re: Re: Soft start
Hi Nigel,
Like jaycee said, the current draw is not constant and also the thermistor get pretty hot. The simple circuit I posted will short the thermistor after the inrush current. The delay is small but notable (I thought it would be less) before the relay closes.
Here's the one I'm using:
I have 2 small relays in parallel (redundant).
nigelwright7557 said:
Hi Nigel,
Like jaycee said, the current draw is not constant and also the thermistor get pretty hot. The simple circuit I posted will short the thermistor after the inrush current. The delay is small but notable (I thought it would be less) before the relay closes.
Here's the one I'm using:
I have 2 small relays in parallel (redundant).
ostripper said:
I already have 2 of the thermisters , mouser and digikey have them as well. You are saying that all I have to do is delay the
relay unit until the thermister's current fully energizes the supply caps ?
(attached)
I want to go "dummy" (simple), until I add a remote controlled
preamp to this project (2 lm3886 rear amps as well , as I have
a 300VA 30-0-30 to use as well). Then I might play with some
of the logic controlled soft starts andy linked to.(fancy on/off
switches)
OS
I use a PIC micro in my amps to delay the speaker relay for 3 seconds on power up.
The PIC also monitors the output volts for a DC condition and if it sees one disconnects the speakers.
No reason why you cant use the same PIC to control the inrush current with a resistor and a relay.
My preference for soft start ckt:
http://www.diyaudio.com/forums/showthread.php?threadid=89576
OS,
A little off topic but I forgot to catch up with your techincs non switching bias scheme investigation. Do you have any conclusion to share?
Cheers.
Atiq
http://www.diyaudio.com/forums/showthread.php?threadid=89576
OS,
A little off topic but I forgot to catch up with your techincs non switching bias scheme investigation. Do you have any conclusion to share?
Cheers.
Atiq
Attachments
Still building.. slow, but perfect.
Yes, I like the circuit.. but it does not fall into the scheme of things.
I am going to trigger my softstart , DC detect , overcurrent from
the same 4017/ optoisolator circuit I have developed.
This will only require that the softstart is buffered by a darlington
to enable turnon delay.
For temp , I use this..
R3 is the relay, In real life, I need the big trannies because I
blew the "magic smoke" out of a 2sc2682 darlington. It was fine
for 5 minutes but the poor little device could not stand the load
(potter brumsfield 5V relay needs 200ma to fully activate), but the
mje runs cold indefinitely. (C1 is temp for delay- r2 will be the
input from the control board.
I have a nice AUX supply (+- 40V and +- 18v) so why not use it ?
(small toroid)...
Still need a big bolt , potted a 2" pvc pipe and a 1" PVC pipe,
they fit perfect in the 1.5KVA and 300VA trafo's.
The wiring was hard , 8 ga to 10ga , split into 2 (for each
40A bridge) , 8 ga CT of trafo split to 2- 10 gauges for main star
ground (glad I made the star with 9 connects) and 10 ga
for DC out to amps.
Also, speakers are doubled 14 ga for ground and yellow 10ga
for OP.Small green wires are 14 ga for each heatsink ( on hinges
makes them a "bad" ground.
Did not have to make any holes on the chassis except for
the toroid bolt.
By tommorrow my new softstart (with overkill darlington) and
LM 317t/337t 12v supply will be stuffed and working.
I will throw the frugal amp 1's I have in the case..
Hope they will survive... I know the PS will..
OS
Yes, I like the circuit.. but it does not fall into the scheme of things.
I am going to trigger my softstart , DC detect , overcurrent from
the same 4017/ optoisolator circuit I have developed.
This will only require that the softstart is buffered by a darlington
to enable turnon delay.
For temp , I use this..
An externally hosted image should be here but it was not working when we last tested it.
R3 is the relay, In real life, I need the big trannies because I
blew the "magic smoke" out of a 2sc2682 darlington. It was fine
for 5 minutes but the poor little device could not stand the load
(potter brumsfield 5V relay needs 200ma to fully activate), but the
mje runs cold indefinitely. (C1 is temp for delay- r2 will be the
input from the control board.
I have a nice AUX supply (+- 40V and +- 18v) so why not use it ?
(small toroid)...
An externally hosted image should be here but it was not working when we last tested it.
Still need a big bolt , potted a 2" pvc pipe and a 1" PVC pipe,
they fit perfect in the 1.5KVA and 300VA trafo's.
The wiring was hard , 8 ga to 10ga , split into 2 (for each
40A bridge) , 8 ga CT of trafo split to 2- 10 gauges for main star
ground (glad I made the star with 9 connects) and 10 ga
for DC out to amps.
An externally hosted image should be here but it was not working when we last tested it.
Also, speakers are doubled 14 ga for ground and yellow 10ga
for OP.Small green wires are 14 ga for each heatsink ( on hinges
makes them a "bad" ground.
Did not have to make any holes on the chassis except for
the toroid bolt.
By tommorrow my new softstart (with overkill darlington) and
LM 317t/337t 12v supply will be stuffed and working.
I will throw the frugal amp 1's I have in the case..
Hope they will survive... I know the PS will.. An externally hosted image should be here but it was not working when we last tested it.
OS
AOK .. ready for a psychoamp..
After 24 hours of "burn in" (large lamps on the 75volt supplies and 15 ohm resistors on the 12v supplies) , all runs smooth
and cool. The softstart relay (at 160ma / 4.9v) does not overheat
the mje15030/2sc1845 darlington and comes on exactly 2 seconds after the mains are engaged.
Board 1 (last post).
A power hog of a relay , but 30A @ 125v (overkill)
The readings are 7.2A.. inrush through the thermister .. 1 second
to +-75v on the main rails , main relay then engages.
Had some fun with the LM317/337. These are real good regulators if properly implemented.
With 10R load (1.2A), only .002v fluctuation (not bad)..
(Board 2 last post) With 6A bridge , 3A toroid secondary, short
circuit protection. The VSOP op-amp front end should be very
happy with this PS (overkill again)
I do not know everything
so if any "groundmasters"
review the following schematic , please tell me of any errors.
(for safety's sake)
One bit of indecision is the hookup of both toriods secondaries to the common star ground , I have seen conflicting opinions
regarding this...? it all works ,no fire
(or even warmth) .
The extra 26-0-26VAC is for the two LM3886 rear channel amps.
For the main amps , I think the VSOP is the way . I prototyped
it , forgot the miller caps and it still did not oscillate ..OMG
OS
After 24 hours of "burn in" (large lamps on the 75volt supplies and 15 ohm resistors on the 12v supplies) , all runs smooth
and cool. The softstart relay (at 160ma / 4.9v) does not overheat
the mje15030/2sc1845 darlington and comes on exactly 2 seconds after the mains are engaged.
An externally hosted image should be here but it was not working when we last tested it.
Board 1 (last post).
A power hog of a relay , but 30A @ 125v (overkill)
The readings are 7.2A.. inrush through the thermister .. 1 second
to +-75v on the main rails , main relay then engages.
Had some fun with the LM317/337. These are real good regulators if properly implemented.
With 10R load (1.2A), only .002v fluctuation (not bad)..
An externally hosted image should be here but it was not working when we last tested it.
(Board 2 last post) With 6A bridge , 3A toroid secondary, short
circuit protection. The VSOP op-amp front end should be very
happy with this PS (overkill again)
I do not know everything
so if any "groundmasters"review the following schematic , please tell me of any errors.
(for safety's sake)
An externally hosted image should be here but it was not working when we last tested it.
One bit of indecision is the hookup of both toriods secondaries to the common star ground , I have seen conflicting opinions
regarding this...? it all works ,no fire
(or even warmth) .The extra 26-0-26VAC is for the two LM3886 rear channel amps.
For the main amps , I think the VSOP is the way . I prototyped
it , forgot the miller caps and it still did not oscillate ..OMG
OS
Attachments
Hi Os
See you have been busy with the monster supply.
Since you have the ksa1381 pair, how is the beta matching between the compliments. I only have 12 pairs of my fav part left and have to wait 5 months for replacements so Im thinking of using these fairchilds for now which are not optimum and will worsen a little problem I will mention later on but maybe do a fair job.
Ive concluded building a veroboard prototype based on the pioneer circuit, the pioneer is a hell of a good amp, my hats off. Im waiting for 5 prototype pc boards from a supplier for further testing on optimum layout. Pioneer werent lying about the specs when they say that amp will do less than .003 THD full power THD20, here we battle doing this 20 years later and with much complexity. This pioneer circuit was in production for over 15 years as the top of range amp, and is still used with some changes in amps only available in japan, they cost little fortunes. Ive made changes to the input section, compensation, dropped the non switching circuitry, no need for that, and with little more complexity 24 trannies exact, .001 THD20 200 W full power is kids play for this thing, I have to multiply the THD20 by nearly 9 to reach that siming it. Ltspice goes bonkers at 1Khz, just shows a flat straight line if I try sim at anything below 120w. Distortion figures at 100 Khz are lower than a lot of amps here at 20khz. Im closer to wire with gain.
It sounds just mighty fine, stumps my electrocompaniet nemos in fine detail and soundstage. Bass is a little down but thats due to the nemos outputstage. It does have some headaches, vas fighting, never in a amp did I have to deal with this but its only apparent now because of the performance of this amp and here I can clearly see it.
Im retiring all my other designs now to the scrap heap. From now on I will only be building this and the Le gecko type amp.
See you have been busy with the monster supply.
Since you have the ksa1381 pair, how is the beta matching between the compliments. I only have 12 pairs of my fav part left and have to wait 5 months for replacements so Im thinking of using these fairchilds for now which are not optimum and will worsen a little problem I will mention later on but maybe do a fair job.
Ive concluded building a veroboard prototype based on the pioneer circuit, the pioneer is a hell of a good amp, my hats off. Im waiting for 5 prototype pc boards from a supplier for further testing on optimum layout. Pioneer werent lying about the specs when they say that amp will do less than .003 THD full power THD20, here we battle doing this 20 years later and with much complexity. This pioneer circuit was in production for over 15 years as the top of range amp, and is still used with some changes in amps only available in japan, they cost little fortunes. Ive made changes to the input section, compensation, dropped the non switching circuitry, no need for that, and with little more complexity 24 trannies exact, .001 THD20 200 W full power is kids play for this thing, I have to multiply the THD20 by nearly 9 to reach that siming it. Ltspice goes bonkers at 1Khz, just shows a flat straight line if I try sim at anything below 120w. Distortion figures at 100 Khz are lower than a lot of amps here at 20khz. Im closer to wire with gain.
It sounds just mighty fine, stumps my electrocompaniet nemos in fine detail and soundstage. Bass is a little down but thats due to the nemos outputstage. It does have some headaches, vas fighting, never in a amp did I have to deal with this but its only apparent now because of the performance of this amp and here I can clearly see it.
Im retiring all my other designs now to the scrap heap. From now on I will only be building this and the Le gecko type amp.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.