[AU] -Get em while/if you can. 0-30V 2.5A P/Supply $49 -DickSmith, last stock!

[mAJORD]

....Teardown incoming 

For anyone after an entry level supply, or maybe another one (who doesn't) - These old supplies are no longer being stocked after this runs out.
They were down to an already decent $99, but recently unknown to me were reduced to $49

Model: Q1770
0-30V  Adjustable Voltage
0-2.5A adjustable current
Ripple:  5mV
Line/Load Regulation: +-10Mv

To those interested in this. I have to apologise first out. I should have posted this as soon as I saw it yesturday arvo, but wanted to look them up and check them out first.. didn't want to post up and have none for me to pickup today 

However.  Unfortunatly, in the meantime, One DAY later DSE have removed the product page from their website! so you can't look up which stores have them in stock anymore.. argghh  (and it's not many)

Thankfully I did a quick look around at the time, and I can only remember the following!.. Sorry!

BRIS: 
North Lakes - 1 in stock (I took # 2)
Logan Hyperdome : Low stock

SYD: One of the western stores came up with Low stock . You'll have to ring up.

MELB: - don't believe anyone had stock.

All the the other stores, who knows. but whatever they have, that's it.
I've only had a quick play, and all seems great. These are a common-ish old unit, and i'm sure theres people here who have one of these. I believe BK and a few others market them aswell.

Will follow up with a teardown shortly though.

[kripton2035]

hum .. cats are always curious he ...? specially new smells ...

[joelby]

I've used one of these before. A perfectly good power supply, and a bargain at $49. Hopefully the local store's still got one in the morning!

[SeanB]

Wish I could get one, but the postage.............

[mAJORD]

hum .. cats are always curious he ...? specially new smells ...

Yes!, Normally out of bounds in the wk/shop as she will play with anything and everything.

Some Inside pic's for the curious people (and cats)












I didn't want to take the front PCB off, as it's hard to get off without removing the heatsink and everything. Not much to see really, all through hole discrete stuff by the looks.

output ripple seems well below specs, even under load. Will put it on the DSO tommorow probably.

It has 2 Taps by the way, swtiches over at 15v

[ciccio]

It looks very well built.
All required insulations are there, wires are of the correct gauge, heatsink is generous...
My impression is that it can give years of service and, in case, be services easily.
Best regards

[Bored@Work]

Probably made by Manson.

Likely one of these ones
http://www.manson.com.hk/en/dcpowersupplies_detail.php?m=3&id=4
http://www.manson.com.hk/en/dcpowersupplies_detail.php?m=3&id=5

[T4P]

Judging from the small knob, it's definitely manson
And apparently the heatsink is nice and we all know that costs some money, but i always like a fan ... cheaper lighter

[joelby]

I just rang up an Adelaide store and the only one left (not sure if this was nationwide or just on the 'local' radar) is in Mildura. Bit of a long drive!

[EEVblog]

IIRC these supplies have (or used to have) nasty overshoot at power on.
Worth checking.

Dave.

[mAJORD]

Will check that out, thanks Dave.

[pickle9000]

Google this one. Look familiar?

BK Precision Model 1670

...mike

[amspire]

I followed the links provided below and read the manual. It says to always turn the supply on first before the equipment to be powered and to turn the equipment off first before the supply.

Is that an indication of power on overshoot?

It doesn't sound good, does it? There is no output switch, so most people will probably carefully connect the circuit up, and then press the ON switch.

The problem with unplugging and plugging in the load is that when you start using multiple supplies, meters, etc, it is very easy to plug the wrong lead into the wrong place. It is better if you can wire everything up, double check all the connections, and then leave everything connected until you finish the work.

If it does have a startup problem, I would work out how to eliminate the problem or add an output switch.

If anyone is checking the supply on a scope, can you check the turn off as well, particularly at no load? If the supply to the regulator circuit dies before the main supply, the output voltage could do anything.

Richard.

[mAJORD]

I followed the links provided below and read the manual. It says to always turn the supply on first before the equipment to be powered and to turn the equipment off first before the supply.

Is that an indication of power on overshoot?

It doesn't sound good, does it? There is no output switch, so most people will probably carefully connect the circuit up, and then press the ON switch.

The problem with unplugging and plugging in the load is that when you start using multiple supplies, meters, etc, it is very easy to plug the wrong lead into the wrong place. It is better if you can wire everything up, double check all the connections, and then leave everything connected until you finish the work.

If it does have a startup problem, I would work out how to eliminate the problem or add an output switch.

If anyone is checking the supply on a scope, can you check the turn off as well, particularly at no load? If the supply to the regulator circuit dies before the main supply, the output voltage could do anything.

Richard.

Sorry I didn't get to capture switchoff, but I could see it just dropped off slowly. Will have another look on the analogue cro here.

I did test for overshoot though with and without load..

5V No Load:


5V , 600mA Load


15.2 V (to check switch on with 2nd tap)


And a 1.4v ,  with around 100mA load.


So no major problems I can see. A bit of noise right at switch on, but minimal overshoot. Perhaps there was an issue at an earlier revision? who knows.


Ripple:

5V  60mA LOAD   3.36mV Pk-Pk
 


16V  2.5A Load 3.84mV Pk-Pk

 
It is a Manson BTW,  I found it printed on one of the PCBs,  Thanks BoredatWork..


-edit, oops, 3v ripple haha

[HackedFridgeMagnet]

I have an old Manson 9615 that has the power on symptoms.
I just thought it was broken and never got around to fixing it and so I always had to make sure that I had critical stuff disconnected during switch on.
It was a bit of a dangerous charade, where I always left one banana plug out during switch on and switch off and also while off.
It's kind of embarrassing now that I never did get around to fixing it. I was thinking of a delayed turn on of a load relay, but it probably has smaller spikes as the taps change anyway.
I only did lose one circuit because of it.
I did a quick measurement on my new Fluke 1587 using the min/max. Turning on to 10.5 volts I got a max of 26v first time and 34v second time.
I have a dim memory of a 60volt spike when I initially discovered this behaviour.
I don't have a storage cro. So no graphs.

[amspire]

I have an old Manson 9615 that has the power on symptoms.
I just thought it was broken and never got around to fixing it and so I always had to make sure that I had critical stuff disconnected during switch on.
It was a bit of a dangerous charade, where I always left one banana plug out during switch on and switch off and also while off.
It's kind of embarrassing now that I never did get around to fixing it. I was thinking of a delayed turn on of a load relay, but it probably has smaller spikes as the taps change anyway.
I only did lose one circuit because of it.
I did a quick measurement on my new Fluke 1587 using the min/max. Turning on to 10.5 volts I got a max of 26v first time and 34v second time.
I have a dim memory of a 60volt spike when I initially discovered this behaviour.
I don't have a storage cro. So no graphs.

A storage cro is not needed. A diode and a 10uF 100V capacitor connected to a multimeter tells you everything you need to know. The discharge time constant is 100 seconds so the peak reading on any meter will capture the shortest of overvoltage transients.

Richard.

[alm]

Assuming the output impedance is (almost) zero ohms so the cap is charged instantly.

[HackedFridgeMagnet]

Not sure why using the cap and diode peak detector would be better than the min/max of the fluke?

I don't suppose anyone has the schematic for an old Manson NP-9615. Seems like the newer ones have fixed this problem.

[EEVblog]

So no major problems I can see. A bit of noise right at switch on, but minimal overshoot. Perhaps there was an issue at an earlier revision? who knows.

Possibly, or maybe even clone versions.
I was fairly sure one I have measured many years ago looked exactly like this.
I remember we fried a proto boards 5V rail because of the overshoot.

Dave.

[amspire]

Not sure why using the cap and diode peak detector would be better than the min/max of the fluke?

The diode acts as a true peak voltage detector. The fluke has a peak reading memory, so it is limited by the measurement speed. How many samples per second can a Fluke multimeter do?

[alm]

It's not about sampling rate, they don't do continuous acquisition (re-arm time is quite long). In fast min max (peak detect) mode the DMM probably use a similar peak detect scheme, though obviously with much smaller caps and a buffered signal. Minimum duration is usually something like 1 ms or 250 us, even though sampling interval is much longer.

[amspire]

It's not about sampling rate, they don't do continuous acquisition (re-arm time is quite long). In fast min max (peak detect) mode the DMM probably use a similar peak detect scheme, though obviously with much smaller caps and a buffered signal. Minimum duration is usually something like 1 ms or 250 us, even though sampling interval is much longer.

A fluke 87 can see peaks down to 250uS. I assume most cheaper multimeters cannot.  Not everyone has a Fluke 87.

A diode and a 100uF capacitor will show peaks faster then this. A lower capacitor would obviously be much better if you have a meter with Gohm impedance. It would be easy to breadboard a simple diode-capacitor-opamp circuit that can easily detect output glitches to less then 1uS.

The point I was making is you do not need digital scopes or $400 multimeters. You can check your power supply for overshoot with a $10 multimeter and 20c worth of parts, and in terms of detecting overshoot, it can outperform the Fluke 87. It is very easy to start thinking that you can only measure something like overshoot with expensive test gear.

But it did remind me of an important point - to check for overshoot with this supply, make sure that the current limit is set to maximum current.

If the current limit is at a low setting, it may slow the startup rise time enough for the regulator circuit to stabilize.

Richard.

[mAJORD]

But it did remind me of an important point - to check for overshoot with this supply, make sure that the current limit is set to maximum current.

If the current limit is at a low setting, it may slow the startup rise time enough for the regulator circuit to stabilize.

Richard.

Just thought i'd update to say I tested as you said today, and same result as above Screenshots, 50-100mV overshoot at most.

The oscillations before voltage ramps up is more interesting.. look at the timebase, they're about a ~1Mhz -15Mhz) . could probably snub it if a concern for low voltage stuff.. the 1N4001 (I think) diode I had floating around the bench at the time couldn't clamp the negative going spike :s





but anyway,  could be a lot better, but both are  not a concern for most things but worth keeping in mind if you're doing really sensitive stuff. Certainly not as bad as that occasional overshoot on the Jaycar unit Dave reviewed!