EEVblog 1767 - Harogic PXN-90 9GHz Spectrum Analyser TEARDOWN

[EEVblog]

Teardown and detailed analysis of the Harogic PXN-90 9GHz Spectrum Analyser.
https://www.harogic.com/product/harogic-px-series-standard-spectrum-analyzer-receiver/

[ballsystemlord]

So what did Dave do in this video?

• Made sure not to break the spectrum analyzer: N
• Photographed both sides of the boards: N
• Showed us datasheets of all the parts used: N
• Showed us the TG board: N
• Removed the can(s) so we could see underneath them: N

Dave, we need your boss to have a few words with you about your work ethic. (ha ha)
More seriously, yeah, I was a bit disappointed, but I know that you're trying.

[EEVblog]

I was a bit disappointed

[pdenisowski]

Thanks Dave - very nicely done (as always). 

I'm very surprised to see a "portable" analyzer that requires tools / disassembly to replace the battery.  I spent many years using portable spec ans outside, all day, and given a typical battery life of 2-4 hours, you absolutely want the ability to (hot) swap batteries when you're doing field work. 

All of portable / handheld spec ans I've ever used from every other T&M manufacturer (and trust me, I've used them all ), allowed you to swap batteries in the field without tools / disassembly.

[tszaboo]

This is very good value for a portable real time analyzer. I was going thorough the Youtube comments, and that's seems to be lost on everyone. Some people even compared it to the TinySA. This is a real time analyzer, not just a swept spectrum one. Some of the built in features are options in A brand analyzers, and the option is going to cost you more than this entire device. The analysis bandwidth by default is 50MHz. Compare that with this:
Rigol RSA5000 real-time analysis 40MHz bandwidth upgrade €1,190.00
Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

[pdenisowski]

Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

Honestly, there's not much of a market for handheld / portable EMI (pre)compliance testing - the ability to move the instrument around isn't very important in that application.

[tszaboo]

Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

Honestly, there's not much of a market for handheld / portable EMI (pre)compliance testing - the ability to move the instrument around isn't very important in that application.

I've seen people use open air labs to do EMC. Meaning that they drive to the middle of the countryside, and do their test there. The other sad reality of today is, that a small engineering office has a limited budget for equipment. When you need to justify the money spent, you cannot really have multiples of specialized boxes. You will get one high speed scope, one SA, one DC load. And hope that the previous purchase decision wasn't a dud.

[temperance]

Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

Honestly, there's not much of a market for handheld / portable EMI (pre)compliance testing - the ability to move the instrument around isn't very important in that application.

There is a huge market for EMI scanning in 2 minutes instead of 30 minutes and sometimes even longer. If it could do that I would buy one.

[pdenisowski]

I've seen people use open air labs to do EMC. Meaning that they drive to the middle of the countryside, and do their test there.

In my experience, even at most OATS (open air test sites), "benchtop" EMC receivers are used

[pdenisowski]

There is a huge market for EMI scanning in 2 minutes instead of 30 minutes and sometimes even longer. If it could do that I would buy one.

Oh yes, time domain scan is HUGE, especially now that (many) standards explicitly allow it.  I suppose if you have a RTSA with 100 MHz real time bandwidth, this could in fact be implemented (since the FFT part is already there).

[mawyatt]

 Dave,

In the design of an RF/MW/MMW systems, there are benefits and tradeoffs wrt the location of the Preselect Filters. These filters have loss (unless cryogenically cooled types) and usually limit the ultimate system sensitivity if located up front at the input port, but help reduce out-of-band signals that aren't of interest (and may cause linearity issues). If moved back in the system and preceded by an LNA their loss can be hidden, however now the LNA is exposed to the out-of-band signals and the linearity consequences. So always trade off!!

In a SA or high DR Wide-Band Heterodyne Rx design the ideal situation is what's called "Mixer First" which basically means that, the 1st Heterodyne mixer is located right up front ahead of everything. Of course this now exposes the mixer to everything at the input (antenna) port!! This is why DARPA called their workshop "Mixer First" way back when Cornell and we presented the PolyPhase (N-Path) Mixer, which created some special unique benefits (like beating the theoretical bi-phase mixer NF, and pushing a narrow baseband filter around the LO) for these types of applications (think EW). This workshop was standing room only, and created a flurry of activity in academia and research facilities, which was the intent

https://www.eevblog.com/forum/rf-microwave/polyphase-or-n-path-mixer/msg3381802/#msg3381802

Anyway, interesting video as usual and thanks for posting

Best

[tszaboo]

Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

Honestly, there's not much of a market for handheld / portable EMI (pre)compliance testing - the ability to move the instrument around isn't very important in that application.

There is a huge market for EMI scanning in 2 minutes instead of 30 minutes and sometimes even longer. If it could do that I would buy one.

At work, we use the Tekbox EMCview and s Rigol RSA 5000 series for this. When we scan from 30MHz up it's really two minutes. AFAIK you only need a few hundred EUR option from Tekbox for the software and the EMI option for the Rigol to be able to do it.
Annoying part is that we don't have an anechoic chamber or a million EUR to buy one. I can get rid of the ambient noise by placing an attenuator, but I don't think the standard allows that (joke). And that there is a relay in the RSA, clicking during the test. I discussed the setup here:
https://www.eevblog.com/forum/testgear/emc-precompliance-testing-setup/msg6011257/#msg6011257

[temperance]

I didn't know you could do this with the RSA 5000 series. Thanks fore the insight.

[nctnico]

Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

Honestly, there's not much of a market for handheld / portable EMI (pre)compliance testing - the ability to move the instrument around isn't very important in that application.

I beg to differ. The first thing I pull out for doing EMC precompliance work is my handheld SA. It is much easier to setup on a bench.

[pdenisowski]

I beg to differ. The first thing I pull out for doing EMC precompliance work is my handheld SA. It is much easier to setup on a bench.

To each his own   But having spent a lot of time with a lot of customers doing precompliance testing, I would have to say that you are very much in the minority

My guess is that when you say "pre-compliance" you mean more "troubleshooting / poking around looking for noise or spurs" rather than "trying to simulate the compliance test environment and conformance to limits in standards", so maybe this is a difference in terminology.

And if anyone is interested, here's my intro to precompliance testing (from InCompliance magazine):  https://incompliancemag.com/pre-compliance-emi-testing/ 

There's also a whitepaper version (attached) and a video version:

 

I do mention the importance of RTSA (or "realtime") in EMC testing - when you're trying to find a spur (especially a pulsing or non-CW one), speed and a good spectrogram can make a big difference 

[nctnico]

I beg to differ. The first thing I pull out for doing EMC precompliance work is my handheld SA. It is much easier to setup on a bench.

To each his own   But having spent a lot of time with a lot of customers doing precompliance testing, I would have to say that you are very much in the minority

My guess is that when you say "pre-compliance" you mean more "troubleshooting / poking around looking for noise or spurs" rather than "trying to simulate the compliance test environment and conformance to limits in standards", so maybe this is a difference in terminology.

In my case both. I do have a TEM cell (large enough for CE / FCC radiated emissions setups) and several types of LISNs so I can get pretty close to actual test results. Together with some software running on a PC a portable SA works just fine for me. Maybe it is just in the heads of people they really need a bulky piece of equipment.

[pdenisowski]

In my case both. I do have a TEM cell (large enough for CE / FCC radiated emissions setups) and several types of LISNs so I can get pretty close to actual test results. Together with some software running on a PC a portable SA works just fine for me. Maybe it is just in the heads of people they really need a bulky piece of equipment.

You're absolutely right that there's no reason someone couldn't use a portable spec an for this.  We (R&S) actually have a "receiver mode" (EMI receiver stepped scan, not real-time receiver) option and quasi-peak on some of our handheld spec ans.  For example: https://www.tequipment.net/Rohde-&-Schwarz/FPH-K43/Options/

But I can tell you with absolute certainty (since I can see the numbers), that we sell orders of magnitude more benchtop units than handheld units to EMC customers.

[thm_w]

EMC customers are going to have a dedicated bench or room setup, so yeah portable would make no sense.

We have large products that you cannot pick up and lift onto a desk, so portable is what makes sense here. But I just use a tinySA which is ~$200. R&S obviously would have no interest in that market segment, or the Harogic segment either.

$1500: https://www.harogic.com/product/harogic-px-series-standard-spectrum-analyzer-receiver/

[tszaboo]

EMC customers are going to have a dedicated bench or room setup, so yeah portable would make no sense.

We have large products that you cannot pick up and lift onto a desk, so portable is what makes sense here. But I just use a tinySA which is ~$200. R&S obviously would have no interest in that market segment, or the Harogic segment either.

$1500: https://www.harogic.com/product/harogic-px-series-standard-spectrum-analyzer-receiver/

I don't think it will bite into R&S's core market. But then it makes quite a few other companies offering completely silly. Who is going to buy a  AIM-TTI when you get 4x the frequency for the same price?

Or what is it going to do to with the sales of B brand manufacturers, like Uni-T or OWON or GW-Instek? It makes their entire portfolio look outdated. The frequency range is higher for the same price, and this one is an RTSA, not a regular one.
I'm having second though, that I bough myself a used Anritsu. Which is in the same price and frequency range. It's a great SA. But from the user manual, I can see that I'm missing out on so many software features, and the additional capabilities.

Looks like Harogic didn't really treat the Frequecy range as "We can charge a lot of money for this". I've seen this, when a company starts treating the pricing as cost+margin, everyone who has an old design, and is using target prices will have a bad day.

[nctnico]

EMC customers are going to have a dedicated bench or room setup, so yeah portable would make no sense.

We have large products that you cannot pick up and lift onto a desk, so portable is what makes sense here. But I just use a tinySA which is ~$200. R&S obviously would have no interest in that market segment, or the Harogic segment either.

$1500: https://www.harogic.com/product/harogic-px-series-standard-spectrum-analyzer-receiver/

That looks like a base price to lure customers in. For modern day EMC you need to go over 6GHz. And for RED / FCC BLE / Wifi pre-compliance having 15GHz of bandwidth is minimum (FCC requires testing little over 22GHz). And the prices for the units you'd actually need aren't listed on Harogic's website. From what I've found online the 20 GHz version (PXE-200) costs around US $10k.

[thm_w]

That looks like a base price to lure customers in. For modern day EMC you need to go over 6GHz. And for RED / FCC BLE / Wifi pre-compliance having 15GHz of bandwidth is minimum (FCC requires testing little over 22GHz). And the prices for the units you'd actually need aren't listed on Harogic's website. From what I've found online the 20 GHz version (PXE-200) costs around US $10k.

I only need ~1GHz for FCC so the 4.5 or 6GHz unit would be fine.
Wifi we just use pre-certified modules/boxes, I don't bother to pre test them.

[EEVblog]

That looks like a base price to lure customers in. For modern day EMC you need to go over 6GHz. And for RED / FCC BLE / Wifi pre-compliance having 15GHz of bandwidth is minimum (FCC requires testing little over 22GHz). And the prices for the units you'd actually need aren't listed on Harogic's website. From what I've found online the 20 GHz version (PXE-200) costs around US $10k.

Nothing has changed in the basics though.
30MHz or 80MHz for conducted emissions
And 1GHz for basic radiated emissions testing.
The $1500 unit is fine for that, just like the 1.5GHz Rigol 815 was the defacto standard for basic low cost EMC pre-compliance testing.

6GHz and 18GHz and higher are a different bracket again.

[EEVblog]

Thanks Dave - very nicely done (as always). 
I'm very surprised to see a "portable" analyzer that requires tools / disassembly to replace the battery.  I spent many years using portable spec ans outside, all day, and given a typical battery life of 2-4 hours, you absolutely want the ability to (hot) swap batteries when you're doing field work. 
All of portable / handheld spec ans I've ever used from every other T&M manufacturer (and trust me, I've used them all ), allowed you to swap batteries in the field without tools / disassembly.

Yeah, bit of a downside. I guess they figure you'd use portable power bank.

[EEVblog]

Honestly, the only thing that I don't see is an EMI option. Maybe they can add that with software. From what I saw with the Rigol, you can get a result very close, within 3dB from a Quasy peak filter with the EMI option RTSA mode. Which also makes a scan 2 minutes instead of 30, very good trade-off. Or maybe they don't want to target that market.

I hadn't looked into that yet, no EMC option is surprising.
I have Rigol's new RSA800 that is replacing the venerable DSA815 that was the cheap go-to option for EMC pre-compliance. It has a paid EMI option.

[pdenisowski]

Yeah, bit of a downside. I guess they figure you'd use portable power bank.

If you're doing "non-stationary" field work (interference hunting, direction finding, radiomonitoring, etc.) you really don't want yet another "box" connected to your kit via a cable.  Every extra external "box" -- even a filter or a preamp connected in-line between the analyzer and antenna -- makes it much harder to do things like climbing onto roofs (or running).  It's also harder to maintain a ...  discrete posture ... if you're hauling a bunch of external accessories around with you.

I'm perpetually amazed at companies that offer solutions for field work* that appear to have been designed by people who have not spent time actually working in the field    Simple things like weight, strap/sling attachment points, screen brightness (including the ability to dim the display almost to zero), battery life, etc. are sometimes just as important as RF performance.

*In fairness, I will note that Harlogic does not appear to currently be marketing this unit for "active" field work.