The problem with Tayda is that, while Chinese sellers on ebay have found loopholes for small orders, a small order (like 5 ICs) in Tayda can quickly become expensive when you add postage, VAT, and customs processing fees. For example, five LM13700 plus postage come to EUR 23 + EUR 5,50 VAT + EUR 3 processing fee, which makes it EUR 6,30 per piece. And you have to wait for at least three weeks for the order to arrive.
I was aware of the issue of chasing obscure parts, so when I decided to go down the synth DIY rabbit hole, I planned accordingly and chose designs that did not use OTAs. But, as you can imagine, at some point I got a little tempted and took a gamble with ebay. It didnāt work, but in the meantime I found reputable sellers from whom I can get an LM13700 faster and at half the price compared to Tayda.
Not that I am planning any LM13700 modules anytime soon!
At these prices, we are better off buying a CEM-type, or other special ICs to build a VCA, VCF or whatever. I donāt know where the insistence on using OTAs comes from. Maybe because we are used to designs from the late 1970s-early 1980s, where OTAs were ubiquitous? I remember Don Lancasterās book on guitar pedals that had OTAs all over them, even in distortion pedals where OTAs were completely unecessaryā¦
In the US, for some reason tax doesnāt show up on Tayda orders, and you can get $3.99 shipping; they say it takes 30 days, but on my last order it actually took 10. So those five chips cost $24.
OTAs are used because theyāre useful. Not everything is a VCA or a VCF, so specialized chips are not the complete answer.
LM13700 is readily available, and cheaper than $4ā¦ as a surface mount package. ($1.30 at Digi-Key.) The problem isnāt that OTAs are going away, itās that THT is going away. DIYers are going to have to face that fact and learn SMD techniques sooner or later.
Letās just say that the US and European countries have very different tax regimes!
There is no way these days that you can import anything in the EU without paying your local VAT (usually 20-25%) plus a flat administration fee (within the EU, you have to pay the VAT of the originating country). And thatās for small orders. Anything over ~EUR20 gets slammed with an additional 10-15% import duties on the value of goods AND the postage fees combined.
I lived in five different EU countries as an adult, and itās the same across the borders. Every order I had, from the US, Australia, Thailand, etc, got taxed upon arrival. The only exceptions are JLCPCB where you can prepay the taxes and save on the administration fee, and tiny packages from China that can fit in your mailbox because the customs offices cannot be bothered with (I can imagine they get overwhelmed as well).
On the other hand, I paid 0 tuition fees studying across three countries, while friends and family have gone through extremely expensive medical care by paying only a tiny fraction of the costs.
I took a gamble and it didnāt work, so I will stick to the local shops or avoid OTAs as much as possible (as per my original plan). Lesson learned! I felt that I should tell others what I learned and resolve the mystery of the non-working VCA.
It may be a while before you get to try them in a circuit, or have suitable test equipment to test the device, by which time a refund deadline has passed.
If you want to waste your money on stuff that is likely faulty or counterfeit, and then enjoy the hassle and delay of (hopefully) getting your money back, from criminals you have paid, feel free. I value my time more than that.
I shop around, plan what I need, and also bulk order commonly used parts for stock. Tayda service is fast, and DHL 2-4 days isnāt too expensive if the postage cost is spread over a lot of items. Mouser are pretty good, CPC, Rapidonline and Farnell are also reliable.
Digikey only seems to want to sell in 10,000+ quantities, for some parts.
Another reason not to buy cheap stuff that is likely to be a fake:
If you are new to electronics, and you have worked diligently to build your first circuit, and you unknowingly put a fake chip inā¦ and it doesnāt work, and you canāt work out what you have (i.e. havenāt) done wrong, it can be pretty demotivating.
On the other hand, the satisfaction of building something you donāt yet fully understandā¦ and it worksā¦ that can really fire your enthusiasm.
Iāve used test circuits present in the datasheets of ICs that you can find via your favourite search engine to run tests on ICs when I got suspicious of them. These test circuits are often very easy to set up. ICs are often produced by several companies who try to adhere to the same specs, so if one datasheet does not show any test circuits, then try and find a different one from another maker.
Iām currently playing the lottery on some lm13700Nās. Last reviewer had tested them and had a photo of his test setup with scope. Once I confirm they are good I might try to make a few test measurements. Currently Iām thinking resistance from Vcc to Vss and diode tester in the Ibias and buffer transistors.
My thinking it might provide people a quick way to verify with a multimeter you have something that is at least LM13700 shaped before using it.
I just received 10 through hole LM13700 which I bought from LCSC. They are genuine parts (tested). They were very well packaged. They cost US$0.41 each. Shipping was very reasonable to Australia. These were the ones I bought: LM13700N HGSEMI | C5160099 - LCSC Electronics
Good to know these work too. I recently bought a bunch of Xinluda XD13700 from LCSC and they also work well in my OTA filters. Havenāt tested them in other circuits yet
I read that the, also obsolete, NE5517 is a pin to pin equivalent. So I looked around in the usual places thinking that there are less chances of faking unknown parts and took another gamble on a seller offering two of those, allegedly made by Philips, for 2something USD. Off to the tester only to find out that one worked but the other didnāt My guess is that these were not fakes but factory rejects that found their way in the market. Synth circuits are typically fine with lower specifications than whatās in the datasheet, so the rejects might work. Or not! The one that tested okay did work fine in the circuit so I kept it, thinking I essentially paid 2 EUR for one IC. Not bad, but not worth the risk and hassle.
I also read that there is a Chinese equivalent in production, the XD13700 that you can order online at a few places, although shipping added considerably to the cost.
So I went by the shop by my office and asked if they had any LM13700 by any chance. Turns out that they did have two different brands in their warehouse. Probably old stock, but they brought them over for me the very next day. For 5 EUR I got one home and it tested all fine.
Lesson confirmed. Buy the cheap non-critical stuff in bulk online (resistors, jacks, knobs, IC sockets, etc), and the critical one-offs from reputable sources, locally if possible, to save time and hassle!
I am building a Basari bass drum module following the schemstics of Sandelinos. It uses an LM13700 for the VCO and VCA. During the building process, I was testing the VCA by applying an 40106 LFO to pin 1 of the LM13700, which is the VCA trigger, but the VCA does not work. I am not able to find any build problems, which makes me think whether the LFO input of 12V could have caused internal problems in the LM13700. I have searched the datasheet, but I have not found the information I was looking for.
Does anybody know what the maximum voltage input of pin 1 (and 16) of an LM13700 is? In other words: have I destroyed that input pin, or should there be another issue in my build, that I havenāt found yet?
Thanks for your replies!
The IABC pin is not a voltage input. Itās a current input with a nominal voltage a diode drop or 2 above the negative supply tail. The chip will break if you put more than 2mA in there if I remember correctly. If you connected it straight to 12V with no current limiting resistor it is likely that the chip is blown.
Thanks, that is too bad, especially as the LM13700 becomes harder to find. Luckily I have some extra LM13700s, so no man overboard. I guess I will finish the build before testing it with a new chip in it.
I found Ray Marstonās Nuts & Volts article āUnderstanding and Using āOTAā Op-Amp ICs part 2ā useful when bodging together my first VCA.
Particularly the bit about limiting current.
Luckily I also had some extra LM13700s.
My guess is that these were not fakes but factory rejects that found their way in the market. Synth circuits are typically fine with lower specifications than whatās in the datasheet, so the rejects might work. Or not! The one that tested okay did work fine in the circuit so I kept it, thinking I essentially paid 2 EUR for one IC. Not bad, but not worth the risk and hassle.
