Gopher cps 3205 power supply. Gopher protocol. What is gopher

Sometimes it seems to me that I do not review, I still get an overview of the power supply.
But in this case, it was originally planned to review the power supply. I had seen it before, so it was even more interesting for me to feel it myself and tell others about my impressions.
Tests, disassembly, analysis, everything is as always and in the same place :)

Sometimes you might think that I'm leaving too good feedback about goods. This is partly due not to my desire to praise the product, but to the fact that I always try to carefully select products for reviews, and some I myself ask to be added to the assortment.
This product just belongs to the category of those that I asked to add, and it is clear that I then decided to write a review :)
But this is so, a lyrical digression, and we will get down to business, i.e. for inspection, testing and other interesting things.

As always, I will hide the unpacking under the spoiler, well, so that it is, but it doesn’t bother anyone :)

Unboxing

In addition to the plastic envelope, there was some kind of soft paper

Which, unfortunately, could not protect its native box, but since the PSU is quite strong, it was not particularly threatened by anything.
On the box there is a small sticker on which, in a nutshell and numbers, it is written what is inside.

Inside, everything is quite simple, the PSU is in a tight package, pieces of foam are inserted on the sides, it sits tightly, does not hang out.

Here's the whole set. In general, the complete set of a laboratory PSU does not imply a bunch of accessories; in general, they gave everything you need.
Power cable
Load connection cable
Instruction
power unit

Instruction in English. I won’t say that it contains a lot of information, but the necessary minimum is definitely present. Although such things are usually bought by people who understand what it is, and often the instructions are needed only at the very beginning and just as a hint.

This power supply is available in a bunch of modifications (in theory, there should be about six).
Three versions with different output voltage and maximum current and two versions with three or four digit current display.
I have a CPS-3205 variant, the maximum output voltage is 32 Volts, the maximum current is 5 Amperes.
In the manual, this is the middle column in the description of the characteristics. I remind you that all photos in the review are clickable.

Also in the kit they gave a cable for connecting to the load. Honestly, the cable is rubbish. When I received this PSU, I naturally wanted to at least try it :)
As a load, I took a 12 volt 45 watt car lamp, but since it just lay on the table, I didn’t turn it on for a long time, the table would still come in handy for me :) But when I finished playing, the wire was not much colder than the lamp. True, it is worth saying that I did not unravel it, but tried it as it is.
The cross section of the cable is completely non-existent, it is better to immediately replace such a cable with something more decent.
But the connectors and crocodiles are quite suitable, so you can only replace the cable itself.

I had no complaints about the power cable, except for the little thing that they accidentally sent me (or I accidentally ordered it) a cable for an American socket :(
The good thing is that an ordinary computer cable is suitable, because this was not a problem, I just took the first one I got out of the box.

Okay, I understand that no one is interested in considering cables, so I'll move on to the power supply.
The first thing that catches your eye is the small size. I think everyone understands that a PSU of such power and such dimensions is usually pulsed.
Switching power supplies have their advantages and disadvantages. If we consider a laboratory power supply, then one of the advantages will be that it has small dimensions with high power, in addition, this power supply has passive cooling (but we will return to this during tests).
Of the minuses, they usually mention elevated level ripple and a large capacity at the output, but we will also return to this.
The dimensions of this PSU are 192x120x58mm, taking into account all the protruding parts.

The layout of the front panel is quite comfortable, although not without flaws. I didn't really like the fact that the output terminals and the power switch are on the back. And while the layout of the output terminals can be adjusted, the rear power switch is not very convenient.
On the front panel, in addition to the controls, there is also an indicator of the set and real output voltage and current. This model uses a four-digit voltage indicator and a three-digit current indicator.
There is a model with a four-digit current indicator, it is a little more convenient, but also more expensive.
In this model, current up to one ampere is displayed in the format xxx.mA, more than 1 Ampere - x.xx.A, the right dot of the indicator informs about the transition to the first measurement mode.

Also on the front panel there are indicators -
Voltage stabilization mode
Current stabilization, this power supply does not have a load shutdown when the set current is exceeded, but goes into limit mode at the set value.
Control lock mode.

The current and voltage are set using an encoder. Rotation - decrease/increase, pressing - the choice of discreteness of change.
There is only one encoder, so there is an adjustment selection switch, to adjust the current or voltage.
At the bottom right there are two buttons, blocking the control and turning on the voltage supply to the output terminals

Everything is simple and modest at the back, two output terminals, an input voltage switch, a power switch and a power connector.
Particular attention should be paid to the position of the power supply switch. If the network is 230 volts and the switch is set to 110 volts, then the consequences will be sad.
The output terminals are quite normal, more than enough for a current of 5 amperes.

The entire body is made of anodized aluminum and is a large heatsink.
This design implies not only good heat dissipation, but also a rather rigid structure, the power supply seems to be monolithic.

There are no cooling fins at the bottom, but there are four small legs. These legs somehow reminded me of the old Soviet drafts, the same in appearance and the same slippery, I would advise you to glue pieces of rubber.
In general, the legs are made of either soft plastic or hard rubber, like a fingernail you can push a little, but they slide on a smooth table.

When you turn on the power supply, it immediately turns on in the - Off mode, there is no automatic voltage supply to the output, which is good in its own way, but it would be even better if you could choose whether to turn on or not after power is applied.
Corrections, it turns out there is a choice to apply voltage to the output or not after switching on.
To do this, hold down the on / off button and turn the encoder until there is an inscription (does not appear immediately) - dON, to return it back, you need to do the same until there is an inscription dOF.

First of all, I decided to check how accurately the power supply voltmeter displays the output voltage, for this I drove it over the entire range from 1 Volt to 32.30 every 5 Volts.
There were complaints only at the edges of the range, at a voltage of 1 Volt and 32.30, at the other control points the voltage was displayed very accurately.

When measuring, I set the appropriate voltage, after which the PSU switched to the output voltage display mode. In order for the power supply unit not to work completely without load (this is a difficult mode for a pulsed power supply unit), I attached a small “load” to the output in the form of a 1 kΩ resistor. By the way, in the photo at the same time you can see how the current measurement works at low values, the output current in mA is equal to the output voltage in volts.
During the measurement process, the resistor burned a little, since 32 volts is a bit too much for it.

After measuring the accuracy of displaying the output voltage, I wanted to check probably the most critical parameter for a switching power supply - the amount of voltage ripple at the output.
I immediately decided to check at the maximum declared current - 5 Amperes and at different voltages, from 5 Volts to the maximum.
The oscilloscope probe was in the 1:1 position and 50mV per division was selected, the input operation mode was AC.
An electronic load was used as a load equivalent.
Outwardly, it looked like this.

At low voltages, I didn’t notice any noticeable ripples at all, except that at a voltage of 10 volts, the oscillogram began to show “signs of life”
At the same time, in the photo you can see how the PSU holds voltage at almost the maximum load current (voltmeter on the left).

An increase in the output voltage to 15 and 20 Volts did not practically change the situation, the ripples remained exactly at the same level as at 10.

I tried it at 25 and 30 volts, I got the impression that I didn’t change anything at all, I even started checking the oscilloscope connection connector, but everything was connected fine.

At the end, I set the output to almost the maximum of what could be a PSU, 32.3 Volts and 5.08 Amps. I didn’t set it anymore, since the electronic load stabilizes the current, and when the PSU switched to current stabilization mode, the output voltage dropped to zero and the load automatically turned off. This is not a breakdown, this behavior is determined by the logic of this bundle.

But I had a feeling that I generally take the same screenshot all the time, it’s good that the waveforms were numbered, otherwise I would be 100% confused.
Not only is there no difference in the ripple level between 10 Volts and 32.3 at all, but also the ripple level is at a very decent level, with the declared 30mV, the PSU actually provides about 10mV, this is more than an excellent result, 3 times higher than the declared characteristics .

Having checked the accuracy of displaying the output voltage and the level of output voltage ripple, I moved on to checking another parameter, the accuracy of displaying the output current.
At low currents, there are small complaints about the measurement accuracy, this is due to the fact that the ADC of the device has a small capacity and at low currents it starts to show with a resolution not of 1 digit, but of the last 3-4 digits.
How this manifests itself can be seen in the photo below.

At higher load currents, everything returns to normal, and at load currents of more than 1 Ampere, the readings are displayed with enviable accuracy.

Readings are displayed sign for sign up to the maximum output current.
The tests were carried out at an output voltage of 15 volts, this voltage was chosen as an approximate middle of the maximum.

Somehow in the comments I saw the question, why a lot of measuring instruments?
Below in the photo is a good example of the simultaneous use of several devices.

The second part of my acquaintance with this power supply was no less interesting to me, and maybe even further than the first. Of course it's a demolition.
Before that, I had already seen the insides in the photo, but it's one thing to see, the second is to feel everything yourself.
The front panel of the device is mounted with decorative screws with an internal hexagon, the rear panel with ordinary screws with a Phillips screwdriver head.

Initially, I took a hex bit and got ready to unscrew the front panel, but in fact, for partial disassembly, it is more convenient to unscrew the back.

We unscrew the four screws securing the back panel and move the bottom towards the back panel.

Here it is, a miracle of engineering thought of Chinese developers. Although practice shows that most of these devices are clones from devices from more famous manufacturers, I will not say anything about this device.
First impression - neat, second - very neat!
The board inside does not take up all the space, there is even free space left, which is good.

There will be quite a lot of photos of the insides further, it may seem to someone that they are superfluous, but to be honest, when preparing the material, I reduced their number as much as I could, but when the thing is assembled well, I want to show it from all sides. Consider this a small weakness of mine :)
And on the other hand, I wanted to show how I would like to see an overview of such a device.

A large number of transformers and chokes are clearly visible.
In general, the device contains 7 hank nodes, I will sign them.
Two input chokes
Electronics power transformer
Isolation transformer for driving high voltage transistors
Power transformer
Inverter output choke
Noise-suppressing output choke.

The assembly is of high quality, I would even say that it is almost at the level of industrial devices.

But of course, there were some small “jambs”, the first thing I noticed was that the wire to the input voltage range switch was dangerously close to the output current measurement shunt. We will deal with it later, but as for the rest of the impression, so far only positive ones, the installation is quite free.
By the way, I measured the drop on the shunt at a current of 5 Amperes, it was exactly 75mV, which gives us a shunt resistance of 15mOhm.

An assembly of four capacitors is installed at the input, 2pcs 100uF and 2pcs 180uF, it is interesting that one pair is made by Capxon, the second by Samxon. I won’t say that this is the height of perfection, but you definitely can’t call these capacitors rootless, it’s quite a middle class.
The choice of such a combination of capacitors is due to the desire of the developers to cram everything into a small case, and the height of the case was the key.
Capacitors for large capacity usually have a large height. I recently tried to pick up capacious and low capacitors like 680x200 Volts, but I haven’t found it yet.
But the developers could well put four at 180, it would only be better.
Capacitors have a real capacitance and in total provide the equivalent of 140uF (100+180)/2.
This capacity is quite sufficient for 160 watts of output power.
By the way, the PSU's own consumption is very small, when the power is turned off and the load is off, it still continues to work for about 10 seconds, powered by capacitors.

Approximately the same combination of manufacturers is used at the output, 2200uF Capxon and 1000uF Samxon.
Upset that the capacitors are designed for 35 volts. In my opinion, this is not enough and it would be more correct to use 50 volt capacitors.
To justify the developers, we can only say that PSUs assembled according to the topology of the monitored one (this is a half-bridge PSU) have less voltage surges and the PSU does not always work at the maximum output voltage.
Also in the photo you can see the noise suppression choke.

Photo of individual nodes.
High-voltage transistors have an insulated case, and in addition, there is an insulating heat-conducting gasket between the case and the transistors, the diode bridge is also screwed to the radiator, although this was not necessary for him.
Special insulators are worn for the output diode assemblies, covering the entire case.
The body of the device is connected to the ground contact of the printed circuit board and, accordingly, to the ground contact of the power connector.
Moreover, when I took it apart further, I saw that they had not forgotten to lay a special washer. And they didn’t even just lay it, but laid it in the right place, between the petal and the body!

Someone will say that there is a washer or an additional insulator. I will answer - any engineer will say that such an approach shows the culture of device production, because if they start saving, then usually from such trifles, then the insulator was considered superfluous, then the washer was not installed, or it was not installed correctly.
Those. Judging even by inspection, I can say that there is production control and it works.

Having unscrewed all the screws holding the elements to the case, I was able to remove the board from the case (it is removed similarly to the bottom).

Now we can move on to an even more detailed examination.
But first, I will remove one of the small "jambs", the wire passing near the shunt.
Above, I wrote that a wire runs near the shunt to the input voltage range selection switch.
In our area, this switch is not only useless, but sometimes harmful, even accidentally switching it will burn out the power supply in a matter of seconds, so I simply remove this switch from the circuit.
This alteration is not suitable for residents of those countries where the mains voltage is 100-120 Volts, they will have to isolate the wire.

BP is being developed very simply. We find where the wires from the switch are connected.

And just solder them :)
Everything, the problem is solved at the global level, no wire - no problem.

The input filter deserves special attention.
I reviewed quite a few different PSUs, but in this case, the input filter is assembled as it should be, you can use it as a sample.

There is a fuse at the input, behind it there is a varistor for protection against increased input voltage and surges. The varistor is rated at 470 volts, this is the standard voltage for varistors in this circuit, there are also 430 volts, but they are less common.
1. You can make a remark, the varistor will not work at a supply voltage of 100-120 Volts and it is more correct to put two of them, but in parallel with high-voltage electrolytes, but we have 230 Volts in the network and the installed option suits us more than completely.
2. An MKP class X2 capacitor is installed at the input, a small two-winding choke is installed nearby.
3. The correct Y1 class capacitors are used in the device, this is what I usually scold when inspecting cheap units. Everything is great here.
4. The power filter uses two two-winding chokes, and a thermistor is also visible to limit the charge current of the capacitors.
For such a power filter, the manufacturer is a big plus, they often use one degree of filtration, here is the correct filter with two double-winding chokes, I have only seen this in branded devices.

Scheme of the input power filter.

1. The inverter uses high-voltage field-effect transistors with a built-in zener diode that protects the gate from voltage surges.
2. A diode bridge is installed at the input, it is designed for 4 Amperes 700 Volts (1000 peak).
3. Output diode assemblies, 16 Amperes 400 Volts, and they are turned on a little unusually, but at the same time correctly. The correctness of the inclusion lies in the fact that they are not paralleled, as is usually the case, and one assembly performs the function of one rectifier diode, the second - of another. This inclusion improves the performance of the diodes, since the diodes located inside the same component are paralleled, on the same chip.
4. They even tried to impregnate the winding knots with varnish, maybe they didn’t impregnate, but they tried to varnish it unambiguously.

Output rectifier and filter circuit. I note the correct inclusion of the noise filter and the current-measuring shunt.

1. PWM controller, well, here is a classic in the form of TL494. The developers went the usual and correct way by using a “hardware” PWM controller that the processor already controls, this provides good accuracy in maintaining voltage and current, as well as a faster response to load changes.
The PWM controller is located on the secondary side and is powered by a separate power supply.
2. All power supply electronics are powered by a separate source, assembled on the basis of TNY274.
I was pleased that the developers used quite common components that can be bought offline, and not Chinese counterparts that can only be bought on eBay or Ali.

I did not draw the circuit diagram, since it would take a completely indecent amount of time, but I drew an approximate block diagram of the device.

There were comments on the printed circuit board. The flux, after mounting some capacitors, is not washed off.
It even surprised me, especially against the backdrop of a very neat assembly. I won't say it's critical, but it's annoying.
Maybe the controller, who monitors compliance with the technical process, went to smoke and at that time my board was being soldered. :(
General photo of the board.

Since the markings of all elements are not visible on the general photo of the board, then I will show several photos of individual nodes, perhaps this will help someone in the repair.
Since field-effect transistors are used in the inverter, and the node itself was originally conceived for bipolar control, the engineers slightly modified the control circuit by adding small transistors. As far as I understand, they are needed for more reliable closing of field-effect transistors.

In this PSU, an isolation transformer is used that differs from the standard one for an ATX type PSU (although the circuit is partially similar). The standard BJT transformer contains more windings, two primary and three secondary, but here it is a pure isolation transformer.
The primary winding is connected to a bridge of four small transistors controlled by the TL494. The transformer has two output windings to drive the top and bottom high voltage transistors.
All this is due to the fact that field-effect transistors are used in the inverter, and this also facilitates the operation of the inverter in a wide range of PWM filling up to complete shutdown.

The TL494 "strapping" unit (right) and some elements of the high-voltage circuit snubber (right).

The output secondary power unit (top) and the signal amplifier unit from the shunt.
The internal electronics power supply supplies two voltages. One feeds the PWM controller, the second - the control and indication electronics.

Oddly enough, but the manufacturer varnished this assembly, I had to wash the varnish off the microcircuit to find out what they used here.
In this case, it is a banal LM358. Its characteristics are far from ideal, but quite sufficient for a specific application, it can be replaced, but I do not think that this will improve anything.

After the inspection, I washed the flux off the PCB a little and covered it with a layer of varnish.
Partly to restore the varnish where I washed it off, and just for general additional protection.

With the inspection of the power board finished, let's move on to the control board.
It is located on the front panel, the flux is also partially not washed off (the controller apparently often left for a smoke, it's time to quit), but in this place it is not critical.

We unscrew the three screws that screw the control board to the front panel.

Yes ... but the indicator could have been put more evenly. I noticed this right away and therefore partly disassembled the entire structure in order to align it. Fortunately, access to the indicator is simple, four screws on the front panel and three screws for fixing the printed circuit board.
By and large, the majority most likely will not notice this, and this definitely does not affect the operation and characteristics of the PSU in any way, but as they say, “the sediment remains”.
There were no more comments, everything seems to be neat.

1. An 8-bit microcontroller is used as a "brain".
2. He controls the indication through a pair of registers, which serve to amplify the current and expand the number of pins.
3. All this “disgrace” is powered by a 3.3 Volt linear stabilizer.
4. Another operational amplifier is installed on this board, and also LM358.

A thermal sensor was also seen on the board. Judging by the instructions, the device has protection against overheating, but in normal operation, I did not succeed in overheating it.

The front panel was examined, nothing more interesting was found on it :)
I remembered about the back panel when I was about to twist everything back.
There are also some details on the back panel :)
But there is also no one necessary and useful part, which for some reason was forgotten, the output fuse. If you connect the battery in reverse polarity to the power supply, then at least the output wires will burn out, it is better to add a fuse, the current is 6-7 Amperes.

A small printed circuit board is installed directly on the output terminals, on which one of the output capacitors is located.
The second printed circuit board is soldered to the power switch contacts. It is noteworthy that not only did they install a separate small board, but the switch also breaks both network wires, which is correct, especially for switching power supplies.

At the same time, I measured the real capacitance of the input filter capacitors.
Don't cheat, everything matches. Above, I gave a calculation why it should be 140 uF, I will repeat it here - 100 uF + 180 uF = 280 uF on capacitors connected in parallel.
Since the capacitors are 200 volts, they are also connected in series, which gives 280/2 = 140 microfarads.

After assembling all this, I decided to check the heating of the elements inside the power supply.
Since I had already measured the rest of the parameters, there were much fewer “participants” in the test.
power unit
Electronic load
Non-contact thermometer
Pen and paper.

The test lasted 1 hour 40 minutes. The temperature was checked at intervals of 20 minutes and a current multiple of 1.25 Amperes.
At the end there was an additional 20 minute run at maximum current and voltage.

All measurement results were summarized in a table.
Let the slightly high initial temperature of the diodes not surprise you, the fact is that the PSU was lying on the table, and above it was a table lamp with a halogen, which warmed up this side of the case a little.
In general, the results are not bad, but an additional run at maximum current showed that at high air temperatures the transformer may overheat. for example, if at 25 degrees it is 88, then at 35 it will already be 98, which is close to critical. I would not recommend running the power supply at maximum power for a long time at high temperatures.
You can slightly improve the situation by laying a piece of heat-conducting rubber between the bottom and the transformer, but pushing the bottom in this state will be very problematic.
All these problems are related to the fact that the PSU has passive cooling, as they say - a double-edged sword.
The temperature of the remaining nodes was within acceptable limits. The case under maximum load warmed up to about 55 degrees, you can hold it in your hands, but it's unpleasant.
At the same time, at the end, I measured the PWM temperature of the electronics power controller (TNY274), it was 67 degrees, which is quite excellent.

As one of the applications, I decided to charge the battery, it is clear that it is possible to use a laboratory PSU as a charger, although it is somehow wrong :)
I connected the battery and ...., received an output overvoltage protection. The fact is that 10 volts were set at the output, and there were noticeably more on the battery and the PSU considered this an emergency, I didn’t even turn on the output.
In order to get out of this error, you need to disconnect the battery, flip the current / voltage selection toggle switch and set the output voltage to be greater than the battery voltage.

That's it, the process has begun, the charge current is set to 1 Ampere, the PSU is operating in current stabilization mode.
By the way, I tried to short the terminals (without a battery, of course) when the output is on, a quiet click and the PSU goes into current stabilization mode.

If you raise the current higher, then the voltage at the battery terminals rises to the set threshold (the battery is old) and the PSU goes into voltage stabilization mode.
You can press the control lock button and leave the battery to charge.

A small comment about the features of management.
For about a week and a half, I used this power supply in normal mode, so I can fully describe the sensations of using this device.
Power switch - convenient if the PSU is on the table, but inconvenient if it is on the shelf.
Output terminals - especially unprincipled, turned on and forgot.
Heating - during normal operation as a feeder for various devices, the heating was small, the largest was only during tests.
Indicator - in the photo it looks worse than in real life, LED indication is convenient for better readability, although I would add a light filter.
Management, well, this is generally a big topic, I'll try to describe it.
At first, the control and indication logic was not very familiar to me, the power supply unit has only two indicators, but for good you need four or five (the fifth one to display power).
Adjustment is both convenient and inconvenient.
It is convenient because in reality you most often adjust only one parameter and one encoder is enough, and switching between modes does not cause problems.
It is inconvenient that if the PSU has left the control mode, then to return it is necessary to make one extra click with the encoder (transfer to the display mode of the set parameter), the second click adjusts.
Well, for the time of adjustment, both indicators go into the display mode of the set value. but there's nothing to be done, the costs of minimalism.
But the most inconvenient thing was that after switching to the output parameter display mode and trying to change the parameters again, the PSU turns on the control mode with a minimum discrete.
For example, I adjusted tenths of a volt, released the handle, the PSU switched to normal mode, and if I later decided to increase / decrease, then I will adjust the hundredths, or I need to click on the encoder. It annoyed me, although you can get used to it. I think the developers could well make the memory of the last mode of operation, it would be more convenient.

So summary.
pros
Pretty good build quality. Well, maybe with the exception of little things like a bad flush of the flux and a crookedly installed indicator, but this does not affect either the reliability or the quality of work.
General thoughtful design.
Quality Components
Good circuitry
Very low ripple
The PSU provides excellent performance over the entire range of output voltages and currents
One of the few PSUs where the manufacturer did not save, maybe almost did not save :)

Minuses
Power switch on the back
No output fuse
There are small flaws in the assembly, described in the pros.
Some inconvenience of management, perhaps a matter of habit.
Thin output wires included, change unambiguously.
A large capacitance at the output, because of this, it is not recommended to connect the element under test to the power supply unit if it has voltage applied to the output (for example, low-power LEDs, etc.), but this is a disadvantage of almost all switching power supplies. first you need to connect the load and only then turn on the voltage supply to the output.

My opinion. A good solid power supply with good technical characteristics.
Of course, you can scold for the flux, etc., but these are trifles against the background of the fact that the power supply is assembled really correctly. The level of output ripple surprised me very pleasantly, I didn't expect it.
Also current and voltage measurement provides good accuracy. For voltage, there are small remarks at the edges of the range, for current only at the beginning, but as for me, the accuracy for a device of this class is quite good.
In general, I liked the PSU, a really interesting and high-quality power supply for a radio amateur. Of course, I can find flaws anywhere, but here the proportion of advantages and disadvantages clearly outweighs the advantages.
In general, that rare case .... when they almost didn’t mess up and made a device with characteristics better than those declared.

I hope that the review was useful and will help you make the right choice.
If you have questions, add-ons will be happy to answer or try some other tests.

The product was provided for writing a review by the store. The review is published in accordance with clause 18 of the Site Rules.